{ "0710.0485.pdf": { "id": "0710.0485v2", "title": "Prediction with expert advice for the Brier game", "year": "2007", "authors": [ "Vladimir Vovk", "Fedor Zhdanov" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/0710.0485v2", "abstract": "We show that the Brier game of prediction is mixable and find the optimal learning rate and substitution function for it. The resulting prediction algorithm is applied to predict results of football and tennis matches. The theoretical performance guarantee turns out to be rather tight on these data sets, especially in the case of the more extensive tennis data." }, "0803.0614.pdf": { "id": "0803.0614v4", "title": "Fitness, chance, and myths: an objective view on soccer results", "year": "2008", "authors": [ "Andreas Heuer", "Oliver Rubner" ], "categories": "physics.data-an, physics.soc-ph", "pdf_url": "https://arxiv.org/pdf/0803.0614v4", "abstract": "We analyze the time series of soccer matches in a model-free way using data for the German soccer league (Bundesliga). We argue that the goal difference is a better measure for the overall fitness of a team than the number of points. It is shown that the time evolution of the table during a season can be interpreted as a random walk with an underlying constant drift. Variations of the overall fitness mainly occur during the summer break but not during a season. The fitness correlation shows a long-time decay on the scale of a quarter century. Some typical soccer myths are analyzed in detail. It is shown that losing but no winning streaks exist. For this analysis ideas from multidimensional NMR experiments have been borrowed. Furthermore, beyond the general home advantage there is no statistically relevant indication of a team-specific home fitness. Based on these insights a framework for a statistical characterization of the results of a soccer league is introduced and some general consequences for the prediction of soccer results are formulated." }, "0807.1759.pdf": { "id": "0807.1759v5", "title": "New Mechanics of Generic Musculo-Skeletal Injury", "year": "2008", "authors": [ "Vladimir G. Ivancevic" ], "categories": "q-bio.TO, q-bio.QM", "pdf_url": "https://arxiv.org/pdf/0807.1759v5", "abstract": "Prediction and prevention of musculo-skeletal injuries is an important aspect of preventive health science. Using as an example a human knee joint, this paper proposes a new coupled-loading-rate hypothesis, which states that a generic cause of any musculo-skeletal injury is a Euclidean jolt, or SE(3)-jolt, an impulsive loading that hits a joint in several coupled degrees-of-freedom simultaneously. Informally, it is a rate-of-change of joint acceleration in all 6-degrees-of-freedom simultaneously, times the corresponding portion of the body mass. In the case of a human knee, this happens when most of the body mass is on one leg with a semi-flexed knee -- and then, caused by some external shock, the knee suddenly `jerks'; this can happen in running, skiing, sports games (e.g., soccer, rugby) and various crashes/impacts. To show this formally, based on the previously defined covariant force law and its application to traumatic brain injury (Ivancevic, 2008), we formulate the coupled Newton--Euler dynamics of human joint motions and derive from it the corresponding coupled SE(3)-jolt dynamics of the joint in case. The SE(3)-jolt is the main cause of two forms of discontinuous joint injury: (i) mild rotational disclinations and (ii) severe translational dislocations. Both the joint disclinations and dislocations, as caused by the SE(3)-jolt, are described using the Cosserat multipolar viscoelastic continuum joint model. Keywords: musculo-skeletal injury, coupled-loading--rate hypothesis, coupled Newton-Euler dynamics, Euclidean jolt dynamics, joint dislocations and disclinations" }, "0909.4555.pdf": { "id": "0909.4555v1", "title": "Soccer matches as experiments: how often does the 'best' team win?", "year": "2009", "authors": [ "G. K. Skinner", "G. H. Freeman" ], "categories": "physics.soc-ph, physics.data-an, stat.AP", "pdf_url": "https://arxiv.org/pdf/0909.4555v1", "abstract": "Models in which the number of goals scored by a team in a soccer match follow a Poisson distribution, or a closely related one, have been widely discussed. We here consider a soccer match as an experiment to assess which of two teams is superior and examine the probability that the outcome of the experiment (match) truly represents the relative abilities of the two teams. Given a final score, it is possible by using a Bayesian approach to quantify the probability that it was or was not the case that 'the best team won'. For typical scores, the probability of a misleading result is significant. Modifying the rules of the game to increase the typical number of goals scored would improve the situation, but a level of confidence that would normally be regarded as satisfactory could not be obtained unless the character of the game was radically changed." }, "0912.2302.pdf": { "id": "0912.2302v1", "title": "Synthesis of supervised classification algorithm using intelligent and statistical tools", "year": "2009", "authors": [ "Ali Douik", "Mourad Moussa Jlassi" ], "categories": "cs.CV, cs.LG", "pdf_url": "https://arxiv.org/pdf/0912.2302v1", "abstract": "A fundamental task in detecting foreground objects in both static and dynamic scenes is to take the best choice of color system representation and the efficient technique for background modeling. We propose in this paper a non-parametric algorithm dedicated to segment and to detect objects in color images issued from a football sports meeting. Indeed segmentation by pixel concern many applications and revealed how the method is robust to detect objects, even in presence of strong shadows and highlights. In the other hand to refine their playing strategy such as in football, handball, volley ball, Rugby..., the coach need to have a maximum of technical-tactics information about the on-going of the game and the players. We propose in this paper a range of algorithms allowing the resolution of many problems appearing in the automated process of team identification, where each player is affected to his corresponding team relying on visual data. The developed system was tested on a match of the Tunisian national competition. This work is prominent for many next computer vision studies as it's detailed in this study." }, "1001.2854.pdf": { "id": "1001.2854v1", "title": "Dynamics of tournaments: the soccer case", "year": "2010", "authors": [ "H. V. Ribeiro", "R. S. Mendes", "L. C. Malacarne", "S. Picoli", "P. A. Santoro" ], "categories": "physics.data-an, physics.soc-ph", "pdf_url": "https://arxiv.org/pdf/1001.2854v1", "abstract": "A random walk-like model is considered to discuss statistical aspects of tournaments. The model is applied to soccer leagues with emphasis on the scores. This competitive system was computationally simulated and the results are compared with empirical data from the English, the German and the Spanish leagues and showed a good agreement with them. The present approach enabled us to characterize a diffusion where the scores are not normally distributed, having a short and asymmetric tail extending towards more positive values. We argue that this non-Gaussian behavior is related with the difference between the teams and with the asymmetry of the scores system. In addition, we compared two tournament systems: the all-play-all and the elimination tournaments." }, "1002.0797.pdf": { "id": "1002.0797v2", "title": "Soccer: is scoring goals a predictable Poissonian process?", "year": "2010", "authors": [ "Andreas Heuer", "Christian Mueller", "Oliver Rubner" ], "categories": "physics.data-an, physics.soc-ph", "pdf_url": "https://arxiv.org/pdf/1002.0797v2", "abstract": "The non-scientific event of a soccer match is analysed on a strictly scientific level. The analysis is based on the recently introduced concept of a team fitness (Eur. Phys. J. B 67, 445, 2009) and requires the use of finite-size scaling. A uniquely defined function is derived which quantitatively predicts the expected average outcome of a soccer match in terms of the fitness of both teams. It is checked whether temporary fitness fluctuations of a team hamper the predictability of a soccer match. To a very good approximation scoring goals during a match can be characterized as independent Poissonian processes with pre-determined expectation values. Minor correlations give rise to an increase of the number of draws. The non-Poissonian overall goal distribution is just a consequence of the fitness distribution among different teams. The limits of predictability of soccer matches are quantified. Our model-free classification of the underlying ingredients determining the outcome of soccer matches can be generalized to different types of sports events." }, "1102.5549.pdf": { "id": "1102.5549v4", "title": "Instant Replay: Investigating statistical Analysis in Sports", "year": "2011", "authors": [ "Gagan Sidhu" ], "categories": "stat.AP, cs.AI, physics.data-an, stat.ML", "pdf_url": "https://arxiv.org/pdf/1102.5549v4", "abstract": "Technology has had an unquestionable impact on the way people watch sports. Along with this technological evolution has come a higher standard to ensure a good viewing experience for the casual sports fan. It can be argued that the pervasion of statistical analysis in sports serves to satiate the fan's desire for detailed sports statistics. The goal of statistical analysis in sports is a simple one: to eliminate subjective analysis. In this paper, we review previous work that attempts to analyze various aspects in sports by using ideas from Markov Chains, Bayesian Inference and Markov Chain Monte Carlo (MCMC) methods. The unifying goal of these works is to achieve an accurate representation of the player's ability, the sport, or the environmental effects on the player's performance. With the prevalence of cheap computation, it is possible that using techniques in Artificial Intelligence could improve the result of statistical analysis in sport. This is best illustrated when evaluating football using Neuro Dynamic Programming, a Control Theory paradigm heavily based on theory in Stochastic processes. The results from this method suggest that statistical analysis in sports may benefit from using ideas from the area of Control Theory or Machine Learning" }, "1102.5728.pdf": { "id": "1102.5728v1", "title": "Named Entity Recognition Using Web Document Corpus", "year": "2011", "authors": [ "Wahiba Ben Abdessalem Karaa" ], "categories": "cs.IR, cs.LG", "pdf_url": "https://arxiv.org/pdf/1102.5728v1", "abstract": "This paper introduces a named entity recognition approach in textual corpus. This Named Entity (NE) can be a named: location, person, organization, date, time, etc., characterized by instances. A NE is found in texts accompanied by contexts: words that are left or right of the NE. The work mainly aims at identifying contexts inducing the NE's nature. As such, The occurrence of the word \"President\" in a text, means that this word or context may be followed by the name of a president as President \"Obama\". Likewise, a word preceded by the string \"footballer\" induces that this is the name of a footballer. NE recognition may be viewed as a classification method, where every word is assigned to a NE class, regarding the context. The aim of this study is then to identify and classify the contexts that are most relevant to recognize a NE, those which are frequently found with the NE. A learning approach using training corpus: web documents, constructed from learning examples is then suggested. Frequency representations and modified tf-idf representations are used to calculate the context weights associated to context frequency, learning example frequency, and document frequency in the corpus." }, "1104.2019.pdf": { "id": "1104.2019v1", "title": "Relative locality and the soccer ball problem", "year": "2011", "authors": [ "Giovanni Amelino-Camelia", "Laurent Freidel", "Jerzy Kowalski-Glikman", "Lee Smolin" ], "categories": "hep-th", "pdf_url": "https://arxiv.org/pdf/1104.2019v1", "abstract": "We consider the behavior of macroscopic bodies within the framework of relative locality, which is a recent proposal for Planck scale modifications of the relativistic dynamics of particles which are described as arising from deformations in the geometry of momentum space. These lead to the addition of non-linear terms to the energy-momentum relations and conservation laws, which are suppressed by powers of ratio between the energy E of the particles involved and the Planck mass M_P. We consider and resolve a common objection against such proposals, which is that, even if the corrections are small for elementary particles in current experiments, they are huge when applied to composite systems such as soccer balls, planets and stars, with energies E_{macro} much larger than M_P. We show that this \"soccer-ball problem\" does not arise within the framework of relative locality, because the non-linear effects for the dynamics of a composite system with N elementary particles appear at most of order E_{macro}/ N M_{P}." }, "1202.3728.pdf": { "id": "1202.3728v1", "title": "Reasoning about RoboCup Soccer Narratives", "year": "2012", "authors": [ "Hannaneh Hajishirzi", "Julia Hockenmaier", "Erik T. Mueller", "Eyal Amir" ], "categories": "cs.AI", "pdf_url": "https://arxiv.org/pdf/1202.3728v1", "abstract": "This paper presents an approach for learning to translate simple narratives, i.e., texts (sequences of sentences) describing dynamic systems, into coherent sequences of events without the need for labeled training data. Our approach incorporates domain knowledge in the form of preconditions and effects of events, and we show that it outperforms state-of-the-art supervised learning systems on the task of reconstructing RoboCup soccer games from their commentaries." }, "1206.6814.pdf": { "id": "1206.6814v1", "title": "An Empirical Comparison of Algorithms for Aggregating Expert Predictions", "year": "2012", "authors": [ "Varsha Dani", "Omid Madani", "David M Pennock", "Sumit Sanghai", "Brian Galebach" ], "categories": "cs.AI, cs.LG", "pdf_url": "https://arxiv.org/pdf/1206.6814v1", "abstract": "Predicting the outcomes of future events is a challenging problem for which a variety of solution methods have been explored and attempted. We present an empirical comparison of a variety of online and offline adaptive algorithms for aggregating experts' predictions of the outcomes of five years of US National Football League games (1319 games) using expert probability elicitations obtained from an Internet contest called ProbabilitySports. We find that it is difficult to improve over simple averaging of the predictions in terms of prediction accuracy, but that there is room for improvement in quadratic loss. Somewhat surprisingly, a Bayesian estimation algorithm which estimates the variance of each expert's prediction exhibits the most consistent superior performance over simple averaging among our collection of algorithms." }, "1207.0700.pdf": { "id": "1207.0700v1", "title": "A statistical view on team handball results: home advantage, team fitness and prediction of match outcomes", "year": "2012", "authors": [ "Jens Smiatek", "Andreas Heuer" ], "categories": "stat.OT, physics.data-an, physics.soc-ph, stat.AP", "pdf_url": "https://arxiv.org/pdf/1207.0700v1", "abstract": "We analyze the results of the German Team Handball Bundesliga for ten seasons in a model-free statistical time series approach. We will show that the home advantage is nearly negligible compared to the total sum of goals. Specific interest has been spent on the time evolution of the team fitness expressed in terms of the goal difference. In contrast to soccer, our results indicate a decay of the team fitness values over a season while the long time correlation behavior over years is nearly comparable. We are able to explain the dominance of a few teams by the large value for the total number of goals in a match. A method for the prediction of match winners is presented in good accuracy with the real results. We analyze the properties of promoted teams and indicate drastic level changes between the Bundesliga and the second league. Our findings reflect in good agreement recent discussions on modern successful attack strategies." }, "1207.4471.pdf": { "id": "1207.4471v1", "title": "How does the past of a soccer match influence its future?", "year": "2012", "authors": [ "Andreas Heuer", "Oliver Rubner" ], "categories": "physics.data-an", "pdf_url": "https://arxiv.org/pdf/1207.4471v1", "abstract": "Scoring goals in a soccer match can be interpreted as a stochastic process. In the most simple description of a soccer match one assumes that scoring goals can be described by a constant goal rate for each team, implying simple Poissonian and Markovian behavior. Here a general framework for the identification of deviations from this behavior is presented. For this endeavor it is essential to formulate an a priori estimate of the expected number of goals per team in a specific match. The analysis scheme is applied to approximately 40 seasons of the German Bundesliga. It is possible to characterize the impact of the previous course of the match on the present match behavior. This allows one to identify interesting generic features about soccer matches and thus to learn about the hidden complexities behind scoring goals." }, "1207.4561.pdf": { "id": "1207.4561v1", "title": "Towards the perfect prediction of soccer matches", "year": "2012", "authors": [ "Andreas Heuer", "Oliver Rubner" ], "categories": "physics.data-an, physics.soc-ph", "pdf_url": "https://arxiv.org/pdf/1207.4561v1", "abstract": "We present a systematic approach to the prediction of soccer matches. First, we show that the information about chances for goals is by far more informative than about the actual results. Second, we present a multivariate regression approach and show how the prediction quality increases with increasing information content. This prediction quality can be explicitly expressed in terms of just two parameters. Third, by disentangling the systematic and random components of soccer matches we can identify the optimum level of predictability. These concepts are exemplified for the German Bundesliga." }, "1207.6430.pdf": { "id": "1207.6430v2", "title": "Optimal Data Collection For Informative Rankings Expose Well-Connected Graphs", "year": "2012", "authors": [ "Braxton Osting", "Christoph Brune", "Stanley J. Osher" ], "categories": "stat.ML, cs.LG, stat.AP", "pdf_url": "https://arxiv.org/pdf/1207.6430v2", "abstract": "Given a graph where vertices represent alternatives and arcs represent pairwise comparison data, the statistical ranking problem is to find a potential function, defined on the vertices, such that the gradient of the potential function agrees with the pairwise comparisons. Our goal in this paper is to develop a method for collecting data for which the least squares estimator for the ranking problem has maximal Fisher information. Our approach, based on experimental design, is to view data collection as a bi-level optimization problem where the inner problem is the ranking problem and the outer problem is to identify data which maximizes the informativeness of the ranking. Under certain assumptions, the data collection problem decouples, reducing to a problem of finding multigraphs with large algebraic connectivity. This reduction of the data collection problem to graph-theoretic questions is one of the primary contributions of this work. As an application, we study the Yahoo! Movie user rating dataset and demonstrate that the addition of a small number of well-chosen pairwise comparisons can significantly increase the Fisher informativeness of the ranking. As another application, we study the 2011-12 NCAA football schedule and propose schedules with the same number of games which are significantly more informative. Using spectral clustering methods to identify highly-connected communities within the division, we argue that the NCAA could improve its notoriously poor rankings by simply scheduling more out-of-conference games." }, "1208.4398.pdf": { "id": "1208.4398v1", "title": "A Unified Approach for Modeling and Recognition of Individual Actions and Group Activities", "year": "2012", "authors": [ "Qiang Qiu", "Rama Chellappa" ], "categories": "cs.CV, stat.ML", "pdf_url": "https://arxiv.org/pdf/1208.4398v1", "abstract": "Recognizing group activities is challenging due to the difficulties in isolating individual entities, finding the respective roles played by the individuals and representing the complex interactions among the participants. Individual actions and group activities in videos can be represented in a common framework as they share the following common feature: both are composed of a set of low-level features describing motions, e.g., optical flow for each pixel or a trajectory for each feature point, according to a set of composition constraints in both temporal and spatial dimensions. In this paper, we present a unified model to assess the similarity between two given individual or group activities. Our approach avoids explicit extraction of individual actors, identifying and representing the inter-person interactions. With the proposed approach, retrieval from a video database can be performed through Query-by-Example; and activities can be recognized by querying videos containing known activities. The suggested video matching process can be performed in an unsupervised manner. We demonstrate the performance of our approach by recognizing a set of human actions and football plays." }, "1210.4854.pdf": { "id": "1210.4854v1", "title": "Semantic Understanding of Professional Soccer Commentaries", "year": "2012", "authors": [ "Hannaneh Hajishirzi", "Mohammad Rastegari", "Ali Farhadi", "Jessica K. Hodgins" ], "categories": "cs.CL, cs.AI", "pdf_url": "https://arxiv.org/pdf/1210.4854v1", "abstract": "This paper presents a novel approach to the problem of semantic parsing via learning the correspondences between complex sentences and rich sets of events. Our main intuition is that correct correspondences tend to occur more frequently. Our model benefits from a discriminative notion of similarity to learn the correspondence between sentence and an event and a ranking machinery that scores the popularity of each correspondence. Our method can discover a group of events (called macro-events) that best describes a sentence. We evaluate our method on our novel dataset of professional soccer commentaries. The empirical results show that our method significantly outperforms the state-of-theart." }, "1211.2719.pdf": { "id": "1211.2719v2", "title": "Quantum Consciousness Soccer Simulator", "year": "2012", "authors": [ "N. B\u00e1tfai" ], "categories": "cs.AI, cs.MA", "pdf_url": "https://arxiv.org/pdf/1211.2719v2", "abstract": "In cognitive sciences it is not uncommon to use various games effectively. For example, in artificial intelligence, the RoboCup initiative was to set up to catalyse research on the field of autonomous agent technology. In this paper, we introduce a similar soccer simulation initiative to try to investigate a model of human consciousness and a notion of reality in the form of a cognitive problem. In addition, for example, the home pitch advantage and the objective role of the supporters could be naturally described and discussed in terms of this new soccer simulation model." }, "1211.3578.pdf": { "id": "1211.3578v2", "title": "Chiral Perturbation Theory and the $\\bar B \\bar B$ Strong Interaction", "year": "2012", "authors": [ "Zhan-Wei Liu", "Ning Li", "Shi-Lin Zhu" ], "categories": "hep-ph, hep-ex, hep-lat, nucl-th", "pdf_url": "https://arxiv.org/pdf/1211.3578v2", "abstract": "We have calculated the potentials of the heavy (charmed or bottomed) pseudoscalar mesons up to $O(\u03b5^2)$ with the heavy meson chiral perturbation theory. We take into account the contributions from the football, triangle, box, and crossed diagrams with the 2$\u03c6$ exchange and one-loop corrections to the contact terms. We notice that the total 2$\u03c6$-exchange potential alone is attractive in the small momentum region in the channel ${\\bar B \\bar B}^{I=1}$, ${\\bar B_s \\bar B_s}^{I=0}$, or ${\\bar B \\bar B_s}^{I=1/2}$, while repulsive in the channel ${\\bar B \\bar B}^{I=0}$. Hopefully the analytical chiral structures of the potentials may be useful in the extrapolation of the heavy meson interaction from lattice QCD simulation." }, "1212.1570.pdf": { "id": "1212.1570v1", "title": "A simple method for decision making in robocup soccer simulation 3d environment", "year": "2012", "authors": [ "Khashayar Niki Maleki", "Mohammad Hadi Valipour", "Sadegh Mokari", "Roohollah Yeylaghi Ashrafi", "Mohammad Reza Jamali", "Caro Lucas" ], "categories": "cs.AI, cs.RO", "pdf_url": "https://arxiv.org/pdf/1212.1570v1", "abstract": "In this paper new hierarchical hybrid fuzzy-crisp methods for decision making and action selection of an agent in soccer simulation 3D environment are presented. First, the skills of an agent are introduced, implemented and classified in two layers, the basicskills and the highlevel skills. In the second layer, a twophase mechanism for decision making is introduced. In phase one, some useful methods are implemented which check the agent's situation for performing required skills. In the next phase, the team str ategy, team for mation, agent's role and the agent's positioning system are introduced. A fuzzy logical approach is employed to recognize the team strategy and further more to tell the player the best position to move. At last, we comprised our implemented algor ithm in the Robocup Soccer Simulation 3D environment and results showed th eefficiency of the introduced methodology." }, "1305.6568.pdf": { "id": "1305.6568v1", "title": "Reinforcement Learning for the Soccer Dribbling Task", "year": "2013", "authors": [ "Arthur Carvalho", "Renato Oliveira" ], "categories": "cs.LG, cs.RO, stat.ML", "pdf_url": "https://arxiv.org/pdf/1305.6568v1", "abstract": "We propose a reinforcement learning solution to the \\emph{soccer dribbling task}, a scenario in which a soccer agent has to go from the beginning to the end of a region keeping possession of the ball, as an adversary attempts to gain possession. While the adversary uses a stationary policy, the dribbler learns the best action to take at each decision point. After defining meaningful variables to represent the state space, and high-level macro-actions to incorporate domain knowledge, we describe our application of the reinforcement learning algorithm \\emph{Sarsa} with CMAC for function approximation. Our experiments show that, after the training period, the dribbler is able to accomplish its task against a strong adversary around 58% of the time." }, "1403.7993.pdf": { "id": "1403.7993v1", "title": "A Markovian model for association football possession and its outcomes", "year": "2014", "authors": [ "Javier L\u00f3pez Pe\u00f1a" ], "categories": "math.PR, math.DS, math.ST", "pdf_url": "https://arxiv.org/pdf/1403.7993v1", "abstract": "We propose a bottom-up approach to the study of possession and its outcomes for association football, based on probabilistic finite state automata with transition probabilities described by a Markov process. We show how even a very simple model yields faithful approximations to the distribution of passing sequences and chances of taking shots for English Premier League teams, which we fit using a whole season of granular game data (380 games). We compare the resulting model with classical top-down distributions traditionally used to describe possessions, showing that the Markov models yield a more accurate asymptotic behaviour." }, "1404.2885.pdf": { "id": "1404.2885v1", "title": "A Networks and Machine Learning Approach to Determine the Best College Coaches of the 20th-21st Centuries", "year": "2014", "authors": [ "Tian-Shun Jiang", "Zachary Polizzi", "Christopher Yuan" ], "categories": "stat.AP, cs.LG, cs.SI", "pdf_url": "https://arxiv.org/pdf/1404.2885v1", "abstract": "Our objective is to find the five best college sports coaches of past century for three different sports. We decided to look at men's basketball, football, and baseball. We wanted to use an approach that could definitively determine team skill from the games played, and then use a machine-learning algorithm to calculate the correct coach skills for each team in a given year. We created a networks-based model to calculate team skill from historical game data. A digraph was created for each year in each sport. Nodes represented teams, and edges represented a game played between two teams. The arrowhead pointed towards the losing team. We calculated the team skill of each graph using a right-hand eigenvector centrality measure. This way, teams that beat good teams will be ranked higher than teams that beat mediocre teams. The eigenvector centrality rankings for most years were well correlated with tournament performance and poll-based rankings. We assumed that the relationship between coach skill $C_s$, player skill $P_s$, and team skill $T_s$ was $C_s \\cdot P_s = T_s$. We then created a function to describe the probability that a given score difference would occur based on player skill and coach skill. We multiplied the probabilities of all edges in the network together to find the probability that the correct network would occur with any given player skill and coach skill matrix. We was able to determine player skill as a function of team skill and coach skill, eliminating the need to optimize two unknown matrices. The top five coaches in each year were noted, and the top coach of all time was calculated by dividing the number of times that coach ranked in the yearly top five by the years said coach had been active." }, "1404.3170.pdf": { "id": "1404.3170v2", "title": "Dynamics of a soccer ball", "year": "2014", "authors": [ "Scott Crass" ], "categories": "math.DS", "pdf_url": "https://arxiv.org/pdf/1404.3170v2", "abstract": "Exploiting the symmetry of the regular icosahedron, Peter Doyle and Curt McMullen constructed a solution to the quintic equation. Their algorithm relied on the dynamics of a certain icosahedral equivariant map for which the icosahedron's twenty face-centers--one of its special orbits--are superattracting periodic points. The current study considers the question of whether there are icosahedrally symmetric maps with superattracting periodic points at a 60-point orbit. The investigation leads to the discovery of two maps whose superattracting sets are configurations of points that are respectively related to a soccer ball and a companion structure. It concludes with a discussion of how a generic 60-point attractor provides for the extraction of all five of the quintic's roots." }, "1407.5093.pdf": { "id": "1407.5093v1", "title": "Classification of Passes in Football Matches using Spatiotemporal Data", "year": "2014", "authors": [ "Michael Horton", "Joachim Gudmundsson", "Sanjay Chawla", "Jo\u00ebl Estephan" ], "categories": "cs.LG, cs.CG", "pdf_url": "https://arxiv.org/pdf/1407.5093v1", "abstract": "A knowledgeable observer of a game of football (soccer) can make a subjective evaluation of the quality of passes made between players during the game. We investigate the problem of producing an automated system to make the same evaluation of passes. We present a model that constructs numerical predictor variables from spatiotemporal match data using feature functions based on methods from computational geometry, and then learns a classification function from labelled examples of the predictor variables. Furthermore, the learned classifiers are analysed to determine if there is a relationship between the complexity of the algorithm that computed the predictor variable and the importance of the variable to the classifier. Experimental results show that we are able to produce a classifier with 85.8% accuracy on classifying passes as Good, OK or Bad, and that the predictor variables computed using complex methods from computational geometry are of moderate importance to the learned classifiers. Finally, we show that the inter-rater agreement on pass classification between the machine classifier and a human observer is of similar magnitude to the agreement between two observers." }, "1407.7891.pdf": { "id": "1407.7891v1", "title": "Planck-scale soccer-ball problem: a case of mistaken identity", "year": "2014", "authors": [ "Giovanni Amelino-Camelia" ], "categories": "gr-qc, hep-th", "pdf_url": "https://arxiv.org/pdf/1407.7891v1", "abstract": "Over the last decade it has been found that nonlinear laws of composition of momenta are predicted by some alternative approaches to \"real\" 4D quantum gravity, and by all formulations of dimensionally-reduced (3D) quantum gravity coupled to matter. The possible relevance for rather different quantum-gravity models has motivated several studies, but this interest is being tempered by concerns that a nonlinear law of addition of momenta might inevitably produce a pathological description of the total momentum of a macroscopic body. I here show that such concerns are unjustified, finding that they are rooted in failure to appreciate the differences between two roles for laws composition of momentum in physics. Previous results relied exclusively on the role of a law of momentum composition in the description of spacetime locality. However, the notion of total momentum of a multi-particle system is not a manifestation of locality, but rather reflects translational invariance. By working within an illustrative example of quantum spacetime I show explicitly that spacetime locality is indeed reflected in a nonlinear law of composition of momenta, but translational invariance still results in an undeformed linear law of addition of momenta building up the total momentum of a multi-particle system." }, "1409.0308.pdf": { "id": "1409.0308v1", "title": "Searching for a Unique Style in Soccer", "year": "2014", "authors": [ "Laszlo Gyarmati", "Haewoon Kwak", "Pablo Rodriguez" ], "categories": "cs.SI, physics.soc-ph", "pdf_url": "https://arxiv.org/pdf/1409.0308v1", "abstract": "Is it possible to have a unique, recognizable style in soccer nowadays? We address this question by proposing a method to quantify the motif characteristics of soccer teams based on their pass networks. We introduce the the concept of \"flow motifs\" to characterize the statistically significant pass sequence patterns. It extends the idea of the network motifs, highly significant subgraphs that usually consists of three or four nodes. The analysis of the motifs in the pass networks allows us to compare and differentiate the styles of different teams. Although most teams tend to apply homogenous style, surprisingly, a unique strategy of soccer exists. Specifically, FC Barcelona's famous tiki-taka does not consist of uncountable random passes but rather has a precise, finely constructed structure." }, "1411.1243.pdf": { "id": "1411.1243v1", "title": "Using Twitter to predict football outcomes", "year": "2014", "authors": [ "Stylianos Kampakis", "Andreas Adamides" ], "categories": "stat.ML, cs.CL, cs.SI", "pdf_url": "https://arxiv.org/pdf/1411.1243v1", "abstract": "Twitter has been proven to be a notable source for predictive modelling on various domains such as the stock market, the dissemination of diseases or sports outcomes. However, such a study has not been conducted in football (soccer) so far. The purpose of this research was to study whether data mined from Twitter can be used for this purpose. We built a set of predictive models for the outcome of football games of the English Premier League for a 3 month period based on tweets and we studied whether these models can overcome predictive models which use only historical data and simple football statistics. Moreover, combined models are constructed using both Twitter and historical data. The final results indicate that data mined from Twitter can indeed be a useful source for predicting games in the Premier League. The final Twitter-based model performs significantly better than chance when measured by Cohen's kappa and is comparable to the model that uses simple statistics and historical data. Combining both models raises the performance higher than it was achieved by each individual model. Thereby, this study provides evidence that Twitter derived features can indeed provide useful information for the prediction of football (soccer) outcomes." }, "1411.6496.pdf": { "id": "1411.6496v1", "title": "Automatic Summarization of Soccer Highlights Using Audio-visual Descriptors", "year": "2014", "authors": [ "Arnau Raventos", "Raul Quijada", "Luis Torres", "Francesc Tarres" ], "categories": "cs.IR", "pdf_url": "https://arxiv.org/pdf/1411.6496v1", "abstract": "Automatic summarization generation of sports video content has been object of great interest for many years. Although semantic descriptions techniques have been proposed, many of the approaches still rely on low-level video descriptors that render quite limited results due to the complexity of the problem and to the low capability of the descriptors to represent semantic content. In this paper, a new approach for automatic highlights summarization generation of soccer videos using audio-visual descriptors is presented. The approach is based on the segmentation of the video sequence into shots that will be further analyzed to determine its relevance and interest. Of special interest in the approach is the use of the audio information that provides additional robustness to the overall performance of the summarization system. For every video shot a set of low and mid level audio-visual descriptors are computed and lately adequately combined in order to obtain different relevance measures based on empirical knowledge rules. The final summary is generated by selecting those shots with highest interest according to the specifications of the user and the results of relevance measures. A variety of results are presented with real soccer video sequences that prove the validity of the approach." }, "1501.06881.pdf": { "id": "1501.06881v2", "title": "On convergence to a football", "year": "2015", "authors": [ "Hao Fang", "Mijia Lai" ], "categories": "math.DG, math.AP", "pdf_url": "https://arxiv.org/pdf/1501.06881v2", "abstract": "We show that spheres of positive constant curvature with $n$ ($n\\geq3$) conic points converge to a sphere of positive constant curvature with two conic points (or called an (American) football) in Gromov-Hausdorff topology when the corresponding singular divisors converge to a critical divisor in the sense of Troyanov. We prove this convergence in two different ways. Geometrically, the convergence follows from Luo-Tian's explicit description of conic spheres as boundaries of convex polytopes in $S^{3}$. Analytically, regarding the conformal factors as the singular solutions to the corresponding PDE, we derive the required a priori estimates and convergence result after proper reparametrization." }, "1502.05886.pdf": { "id": "1502.05886v1", "title": "On predictability of rare events leveraging social media: a machine learning perspective", "year": "2015", "authors": [ "Lei Le", "Emilio Ferrara", "Alessandro Flammini" ], "categories": "cs.SI, cs.LG, physics.data-an, physics.soc-ph", "pdf_url": "https://arxiv.org/pdf/1502.05886v1", "abstract": "Information extracted from social media streams has been leveraged to forecast the outcome of a large number of real-world events, from political elections to stock market fluctuations. An increasing amount of studies demonstrates how the analysis of social media conversations provides cheap access to the wisdom of the crowd. However, extents and contexts in which such forecasting power can be effectively leveraged are still unverified at least in a systematic way. It is also unclear how social-media-based predictions compare to those based on alternative information sources. To address these issues, here we develop a machine learning framework that leverages social media streams to automatically identify and predict the outcomes of soccer matches. We focus in particular on matches in which at least one of the possible outcomes is deemed as highly unlikely by professional bookmakers. We argue that sport events offer a systematic approach for testing the predictive power of social media, and allow to compare such power against the rigorous baselines set by external sources. Despite such strict baselines, our framework yields above 8% marginal profit when used to inform simple betting strategies. The system is based on real-time sentiment analysis and exploits data collected immediately before the games, allowing for informed bets. We discuss the rationale behind our approach, describe the learning framework, its prediction performance and the return it provides as compared to a set of betting strategies. To test our framework we use both historical Twitter data from the 2014 FIFA World Cup games, and real-time Twitter data collected by monitoring the conversations about all soccer matches of four major European tournaments (FA Premier League, Serie A, La Liga, and Bundesliga), and the 2014 UEFA Champions League, during the period between Oct. 25th 2014 and Nov. 26th 2014." }, "1503.03948.pdf": { "id": "1503.03948v2", "title": "In vivo evaluation of wearable head impact sensors", "year": "2015", "authors": [ "Lyndia C. Wu", "Vaibhav Nangia", "Kevin Bui", "Bradley Hammoor", "Mehmet Kurt", "Fidel Hernandez" ], "categories": "q-bio.QM, physics.med-ph", "pdf_url": "https://arxiv.org/pdf/1503.03948v2", "abstract": "Inertial sensors are commonly used to measure human head motion. Some sensors have been validated with dummy or cadaver experiments, but methods to evaluate sensors in vivo are lacking. Here we present an in vivo method using high speed video to evaluate teeth-mounted (mouthguard), soft tissue-mounted (skin patch), and headgear-mounted (skull cap) sensors during 6-13g sagittal soccer head impacts. Sensor coupling to the skull is quantified by displacement from an ear-canal reference. Mouthguard displacements were within video measurement error (<1mm), while the skin patch and skull cap displaced up to 4mm and 13mm from the ear-canal reference, respectively. We used the mouthguard, which had the least displacement from skull, as the reference to assess 6-degree-of-freedom skin patch and skull cap measurements. Linear and rotational acceleration magnitudes were over-predicted by both the skin patch (with 120% NRMS error for a_mag, 290% for alpha_mag) and the skull cap (320% NRMS error for a_mag, 500% for alpha_mag). Such over-predictions were largely due to out-of-plane motion. To model sensor error, we found that in-plane acceleration peaks from the skin patch in the anterior-posterior direction could be modeled by an underdamped viscoelastic system. In summary, the mouthguard showed tighter skull coupling in vivo than the other sensors. Furthermore, the in vivo methods presented are valuable for investigating skull acceleration sensor technologies." }, "1508.02171.pdf": { "id": "1508.02171v1", "title": "Automatic Extraction of the Passing Strategies of Soccer Teams", "year": "2015", "authors": [ "Laszlo Gyarmati", "Xavier Anguera" ], "categories": "cs.CV, stat.ML", "pdf_url": "https://arxiv.org/pdf/1508.02171v1", "abstract": "Technology offers new ways to measure the locations of the players and of the ball in sports. This translates to the trajectories the ball takes on the field as a result of the tactics the team applies. The challenge professionals in soccer are facing is to take the reverse path: given the trajectories of the ball is it possible to infer the underlying strategy/tactic of a team? We propose a method based on Dynamic Time Warping to reveal the tactics of a team through the analysis of repeating series of events. Based on the analysis of an entire season, we derive insights such as passing strategies for maintaining ball possession or counter attacks, and passing styles with a focus on the team or on the capabilities of the individual players." }, "1511.06262.pdf": { "id": "1511.06262v1", "title": "Seasonal Linear Predictivity in National Football Championships", "year": "2015", "authors": [ "Giuseppe Jurman" ], "categories": "stat.AP", "pdf_url": "https://arxiv.org/pdf/1511.06262v1", "abstract": "Predicting the results of sport matches and competitions is an arising research field, benefiting from the growing amount of available data and the novel data analytics techniques. Excellent forecasts can be achieved by advanced machine learning methods applied to detailed historical data, especially in very popular sports such as football (soccer). Here we show that, despite the large number of confounding factors, the results of a football team in longer competitions (e.g., a national league) follow a basically linear trend useful for predictive purposes, too. In support of this claim, we present a set of experiments of linear regression on a database collecting the yearly results of 707 teams playing in 22 divisions from 11 countries, in 20 football seasons." }, "1603.05583.pdf": { "id": "1603.05583v1", "title": "Analyzing In-Game Movements of Soccer Players at Scale", "year": "2016", "authors": [ "Laszlo Gyarmati", "Mohamed Hefeeda" ], "categories": "cs.OH, stat.AP", "pdf_url": "https://arxiv.org/pdf/1603.05583v1", "abstract": "It is challenging to get access to datasets related to the physical performance of soccer players. The teams consider such information highly confidential, especially if it covers in-game performance.Hence, most of the analysis and evaluation of the players' performance do not contain much information on the physical aspect of the game, creating a blindspot in performance analysis. We propose a novel method to solve this issue by deriving movement characteristics of soccer players. We use event-based datasets from data provider companies covering 50+ soccer leagues allowing us to analyze the movement profiles of potentially tens of thousands of players without any major investment. Our methodology does not require expensive, dedicated player tracking system deployed in the stadium. We also compute the similarity of the players based on their movement characteristics and as such identify potential candidates who may be able to replace a given player. Finally, we quantify the uniqueness and consistency of players in terms of their in-game movements. Our study is the first of its kind that focuses on the movements of soccer players at scale, while it derives novel, actionable insights for the soccer industry from event-based datasets." }, "1604.02715.pdf": { "id": "1604.02715v1", "title": "Soccer Field Localization from a Single Image", "year": "2016", "authors": [ "Namdar Homayounfar", "Sanja Fidler", "Raquel Urtasun" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/1604.02715v1", "abstract": "In this work, we propose a novel way of efficiently localizing a soccer field from a single broadcast image of the game. Related work in this area relies on manually annotating a few key frames and extending the localization to similar images, or installing fixed specialized cameras in the stadium from which the layout of the field can be obtained. In contrast, we formulate this problem as a branch and bound inference in a Markov random field where an energy function is defined in terms of field cues such as grass, lines and circles. Moreover, our approach is fully automatic and depends only on single images from the broadcast video of the game. We demonstrate the effectiveness of our method by applying it to various games and obtain promising results. Finally, we posit that our approach can be applied easily to other sports such as hockey and basketball." }, "1604.05016.pdf": { "id": "1604.05016v2", "title": "Status Report of the GERDA Phase II Startup", "year": "2016", "authors": [ "Valerio D'Andrea" ], "categories": "physics.ins-det", "pdf_url": "https://arxiv.org/pdf/1604.05016v2", "abstract": "The GERmanium Detector Array GERDA experiment, located at the Laboratori Nazionali del Gran Sasso (LNGS) of INFN, searches for $0\u03bd\u03b2\u03b2$ of $^{76}$Ge. Germanium diodes enriched to $\\sim 86~\\%$ in the double beta emitter $^{76}$Ge ($^{enr}$Ge) are exposed being both source and detectors of $0\u03bd\u03b2\u03b2$ decay. Neutrinoless double beta decay is considered a powerful probe to address still open issues in the neutrino sector of the (beyond) Standard Model of particle Physics. Since 2013, just after the completion of the first part of its experimental program (Phase I), the GERDA setup has been upgraded to perform its next step in the $0\u03bd\u03b2\u03b2$ searches (Phase II). Phase II aims to reach a sensitivity to the $0\u03bd\u03b2\u03b2$ decay half-life larger than $10^{26}~$yr in about 3 years of physics data taking. This exposing a detector mass of about $35~$kg of $^{enr}$Ge and with a background index of about $10^{-3}~$cts/(keV$\\cdot$kg$\\cdot$yr). One of the main new implementations is the liquid argon scintillation light read-out, to veto those events that only partially deposit their energy both in Ge and in the surrounding LAr. In this paper the GERDA Phase II expected goals, the upgrade work and few selected features from the 2015 commissioning and 2016 calibration runs will be presented. The main Phase I achievements will be also reviewed." }, "1604.05790.pdf": { "id": "1604.05790v1", "title": "Position Prediction of Ball and Fuzzy Controller for Shooting Action in A Soccer Robot System", "year": "2016", "authors": [ "GyongIl Ryang", "MyongSong Choe", "YongChol Sin" ], "categories": "cs.RO", "pdf_url": "https://arxiv.org/pdf/1604.05790v1", "abstract": "The robot soccer game based complex motion control has been widely studied for the moving object capture and shooting. A position prediction algorithm based on global vision is introduced in order to improve the accuracy and robustness of the vision system for tracking moving objects, including a Kalmanfiter, a dynamic window and an obstacle avoidance strategy. This paper deals with the positon prediction for moving ball by using Kalmanfiter and the Fuzzy Controller for shooting action in a dynamic environment." }, "1606.00285.pdf": { "id": "1606.00285v1", "title": "Using Monte Carlo Search With Data Aggregation to Improve Robot Soccer Policies", "year": "2016", "authors": [ "Francesco Riccio", "Roberto Capobianco", "Daniele Nardi" ], "categories": "cs.RO", "pdf_url": "https://arxiv.org/pdf/1606.00285v1", "abstract": "RoboCup soccer competitions are considered among the most challenging multi-robot adversarial environments, due to their high dynamism and the partial observability of the environment. In this paper we introduce a method based on a combination of Monte Carlo search and data aggregation (MCSDA) to adapt discrete-action soccer policies for a defender robot to the strategy of the opponent team. By exploiting a simple representation of the domain, a supervised learning algorithm is trained over an initial collection of data consisting of several simulations of human expert policies. Monte Carlo policy rollouts are then generated and aggregated to previous data to improve the learned policy over multiple epochs and games. The proposed approach has been extensively tested both on a soccer-dedicated simulator and on real robots. Using this method, our learning robot soccer team achieves an improvement in ball interceptions, as well as a reduction in the number of opponents' goals. Together with a better performance, an overall more efficient positioning of the whole team within the field is achieved." }, "1606.02253.pdf": { "id": "1606.02253v2", "title": "How to flatten a soccer ball", "year": "2016", "authors": [ "Kaie Kubjas", "Pablo A. Parrilo", "Bernd Sturmfels" ], "categories": "math.AG, math.OC", "pdf_url": "https://arxiv.org/pdf/1606.02253v2", "abstract": "This is an experimental case study in real algebraic geometry, aimed at computing the image of a semialgebraic subset of 3-space under a polynomial map into the plane. For general instances, the boundary of the image is given by two highly singular curves. We determine these curves and show how they demarcate the \"flattened soccer ball\". We explore cylindrical algebraic decompositions, by working through concrete examples. Maps onto convex polygons and connections to convex optimization are also discussed." }, "1608.03532.pdf": { "id": "1608.03532v1", "title": "QPass: a Merit-based Evaluation of Soccer Passes", "year": "2016", "authors": [ "Laszlo Gyarmati", "Rade Stanojevic" ], "categories": "cs.AI, stat.AP, stat.ML", "pdf_url": "https://arxiv.org/pdf/1608.03532v1", "abstract": "Quantitative analysis of soccer players' passing ability focuses on descriptive statistics without considering the players' real contribution to the passing and ball possession strategy of their team. Which player is able to help the build-up of an attack, or to maintain the possession of the ball? We introduce a novel methodology called QPass to answer questions like these quantitatively. Based on the analysis of an entire season, we rank the players based on the intrinsic value of their passes using QPass. We derive an album of pass trajectories for different gaming styles. Our methodology reveals a quite counterintuitive paradigm: losing the ball possession could lead to better chances to win a game." }, "1609.01176.pdf": { "id": "1609.01176v1", "title": "The Player Kernel: Learning Team Strengths Based on Implicit Player Contributions", "year": "2016", "authors": [ "Lucas Maystre", "Victor Kristof", "Antonio J. Gonz\u00e1lez Ferrer", "Matthias Grossglauser" ], "categories": "cs.LG, stat.AP", "pdf_url": "https://arxiv.org/pdf/1609.01176v1", "abstract": "In this work, we draw attention to a connection between skill-based models of game outcomes and Gaussian process classification models. The Gaussian process perspective enables a) a principled way of dealing with uncertainty and b) rich models, specified through kernel functions. Using this connection, we tackle the problem of predicting outcomes of football matches between national teams. We develop a player kernel that relates any two football matches through the players lined up on the field. This makes it possible to share knowledge gained from observing matches between clubs (available in large quantities) and matches between national teams (available only in limited quantities). We evaluate our approach on the Euro 2008, 2012 and 2016 final tournaments." }, "1609.05559.pdf": { "id": "1609.05559v1", "title": "Opponent Modeling in Deep Reinforcement Learning", "year": "2016", "authors": [ "He He", "Jordan Boyd-Graber", "Kevin Kwok", "Hal Daum\u00e9" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/1609.05559v1", "abstract": "Opponent modeling is necessary in multi-agent settings where secondary agents with competing goals also adapt their strategies, yet it remains challenging because strategies interact with each other and change. Most previous work focuses on developing probabilistic models or parameterized strategies for specific applications. Inspired by the recent success of deep reinforcement learning, we present neural-based models that jointly learn a policy and the behavior of opponents. Instead of explicitly predicting the opponent's action, we encode observation of the opponents into a deep Q-Network (DQN); however, we retain explicit modeling (if desired) using multitasking. By using a Mixture-of-Experts architecture, our model automatically discovers different strategy patterns of opponents without extra supervision. We evaluate our models on a simulated soccer game and a popular trivia game, showing superior performance over DQN and its variants." }, "1609.07480.pdf": { "id": "1609.07480v1", "title": "Predictive modelling of football injuries", "year": "2016", "authors": [ "Stylianos Kampakis" ], "categories": "stat.AP, cs.AI, stat.ML", "pdf_url": "https://arxiv.org/pdf/1609.07480v1", "abstract": "The goal of this thesis is to investigate the potential of predictive modelling for football injuries. This work was conducted in close collaboration with Tottenham Hotspurs FC (THFC), the PGA European tour and the participation of Wolverhampton Wanderers (WW). Three investigations were conducted: 1. Predicting the recovery time of football injuries using the UEFA injury recordings: The UEFA recordings is a common standard for recording injuries in professional football. For this investigation, three datasets of UEFA injury recordings were available. Different machine learning algorithms were used in order to build a predictive model. The performance of the machine learning models is then improved by using feature selection conducted through correlation-based subset feature selection and random forests. 2. Predicting injuries in professional football using exposure records: The relationship between exposure (in training hours and match hours) in professional football athletes and injury incidence was studied. A common problem in football is understanding how the training schedule of an athlete can affect the chance of him getting injured. The task was to predict the number of days a player can train before he gets injured. 3. Predicting intrinsic injury incidence using in-training GPS measurements: A significant percentage of football injuries can be attributed to overtraining and fatigue. GPS data collected during training sessions might provide indicators of fatigue, or might be used to detect very intense training sessions which can lead to overtraining. This research used GPS data gathered during training sessions of the first team of THFC, in order to predict whether an injury would take place during a week." }, "1610.00825.pdf": { "id": "1610.00825v1", "title": "Predict genome-scale fluxes based solely on enzyme abundance by a novel Hyper-Cube Shrink Algorithm", "year": "2016", "authors": [ "Zhengwei Xie", "Tianyu Zhang", "Qi Ouyang" ], "categories": "q-bio.MN, q-bio.QM", "pdf_url": "https://arxiv.org/pdf/1610.00825v1", "abstract": "One of the long-expected goals of genome-scale metabolic modeling is to evaluate the influence of the perturbed enzymes to the flux distribution. Both ordinary differential equation (ODE) models and the constraint-based models, like Flux balance analysis (FBA), lack of the room of performing metabolic control analysis (MCA) for large-scale networks. In this study, we developed a Hyper-Cube Shrink Algorithm (HCSA) to incorporate the enzymatic properties to the FBA model by introducing a pseudo reaction constrained by enzymatic parameters. Our algorithm was able to handle not only prediction of knockout strains but also strains with quantitative adjustment of expression level or activity. We first demonstrate the concept by applying HCSA to a simplest three-node network. Then we validate its prediction by comparing with ODE and with a synthetic network in Saccharomyces cerevisiae producing voilacein and analogues. Finally we show its capability of predicting the flux distribution in genome-scale networks by applying it to the sporulation in yeast." }, "1702.00585.pdf": { "id": "1702.00585v1", "title": "The temporalized Massey's method", "year": "2017", "authors": [ "Massimo Franceschet", "Enrico Bozzo" ], "categories": "cs.SI, physics.soc-ph", "pdf_url": "https://arxiv.org/pdf/1702.00585v1", "abstract": "We propose and throughly investigate a temporalized version of the popular Massey's technique for rating actors in sport competitions. The method can be described as a dynamic temporal process in which team ratings are updated at every match according to their performance during the match and the strength of the opponent team. Using the Italian soccer dataset, we empirically show that the method has a good foresight prediction accuracy." }, "1702.05662.pdf": { "id": "1702.05662v4", "title": "Spatial modeling of shot conversion in soccer to single out goalscoring ability", "year": "2017", "authors": [ "Soudeep Deb", "Debangan Dey" ], "categories": "stat.AP", "pdf_url": "https://arxiv.org/pdf/1702.05662v4", "abstract": "Goals are results of pin-point shots and it is a pivotal decision in soccer when, how and where to shoot. The main contribution of this study is two-fold. At first, after showing that there exists high spatial correlation in the data of shots across games, we introduce a spatial process in the error structure to model the probability of conversion from a shot depending on positional and situational covariates. The model is developed using a full Bayesian framework. Secondly, based on the proposed model, we define two new measures that can appropriately quantify the impact of an individual in soccer, by evaluating the positioning senses and shooting abilities of the players. As a practical application, the method is implemented on Major League Soccer data from 2016/17 season." }, "1703.03385.pdf": { "id": "1703.03385v1", "title": "Visual-Interactive Similarity Search for Complex Objects by Example of Soccer Player Analysis", "year": "2017", "authors": [ "J\u00fcrgen Bernard", "Christian Ritter", "David Sessler", "Matthias Zeppelzauer", "J\u00f6rn Kohlhammer", "Dieter Fellner" ], "categories": "cs.LG, cs.IR", "pdf_url": "https://arxiv.org/pdf/1703.03385v1", "abstract": "The definition of similarity is a key prerequisite when analyzing complex data types in data mining, information retrieval, or machine learning. However, the meaningful definition is often hampered by the complexity of data objects and particularly by different notions of subjective similarity latent in targeted user groups. Taking the example of soccer players, we present a visual-interactive system that learns users' mental models of similarity. In a visual-interactive interface, users are able to label pairs of soccer players with respect to their subjective notion of similarity. Our proposed similarity model automatically learns the respective concept of similarity using an active learning strategy. A visual-interactive retrieval technique is provided to validate the model and to execute downstream retrieval tasks for soccer player analysis. The applicability of the approach is demonstrated in different evaluation strategies, including usage scenarions and cross-validation tests." }, "1703.04115.pdf": { "id": "1703.04115v2", "title": "BetaRun Soccer Simulation League Team: Variety, Complexity, and Learning", "year": "2017", "authors": [ "Olivia Michael", "Oliver Obst" ], "categories": "cs.AI", "pdf_url": "https://arxiv.org/pdf/1703.04115v2", "abstract": "RoboCup offers a set of benchmark problems for Artificial Intelligence in form of official world championships since 1997. The most tactical advanced and richest in terms of behavioural complexity of these is the 2D Soccer Simulation League, a simulated robotic soccer competition. BetaRun is a new attempt combining both machine learning and manual programming approaches, with the ultimate goal to arrive at a team that is trained entirely from observing and playing games, and a new development based on agent2D." }, "1705.00885.pdf": { "id": "1705.00885v3", "title": "Quantifying the relation between performance and success in soccer", "year": "2017", "authors": [ "Luca Pappalardo", "Paolo Cintia" ], "categories": "stat.AP, stat.ML", "pdf_url": "https://arxiv.org/pdf/1705.00885v3", "abstract": "The availability of massive data about sports activities offers nowadays the opportunity to quantify the relation between performance and success. In this study, we analyze more than 6,000 games and 10 million events in six European leagues and investigate this relation in soccer competitions. We discover that a team's position in a competition's final ranking is significantly related to its typical performance, as described by a set of technical features extracted from the soccer data. Moreover we find that, while victory and defeats can be explained by the team's performance during a game, it is difficult to detect draws by using a machine learning approach. We then simulate the outcomes of an entire season of each league only relying on technical data, i.e. excluding the goals scored, exploiting a machine learning model trained on data from past seasons. The simulation produces a team ranking (the PC ranking) which is close to the actual ranking, suggesting that a complex systems' view on soccer has the potential of revealing hidden patterns regarding the relation between performance and success." }, "1705.04356.pdf": { "id": "1705.04356v1", "title": "Comparing probabilistic predictive models applied to football", "year": "2017", "authors": [ "Marcio A. Diniz", "Rafael Izbicki", "Danilo Lopes", "Luis Ernesto Salasar" ], "categories": "stat.AP", "pdf_url": "https://arxiv.org/pdf/1705.04356v1", "abstract": "We propose two Bayesian multinomial-Dirichlet models to predict the final outcome of football (soccer) matches and compare them to three well-known models regarding their predictive power. All the models predicted the full-time results of 1710 matches of the first division of the Brazilian football championship and the comparison used three proper scoring rules, the proportion of errors and a calibration assessment. We also provide a goodness of fit measure. Our results show that multinomial-Dirichlet models are not only competitive with standard approaches, but they are also well calibrated and present reasonable goodness of fit." }, "1705.08079.pdf": { "id": "1705.08079v2", "title": "Effective injury forecasting in soccer with GPS training data and machine learning", "year": "2017", "authors": [ "Alessio Rossi", "Luca Pappalardo", "Paolo Cintia", "Marcello Iaia", "Javier Fernandez", "Daniel Medina" ], "categories": "stat.ML, stat.AP", "pdf_url": "https://arxiv.org/pdf/1705.08079v2", "abstract": "Injuries have a great impact on professional soccer, due to their large influence on team performance and the considerable costs of rehabilitation for players. Existing studies in the literature provide just a preliminary understanding of which factors mostly affect injury risk, while an evaluation of the potential of statistical models in forecasting injuries is still missing. In this paper, we propose a multi-dimensional approach to injury forecasting in professional soccer that is based on GPS measurements and machine learning. By using GPS tracking technology, we collect data describing the training workload of players in a professional soccer club during a season. We then construct an injury forecaster and show that it is both accurate and interpretable by providing a set of case studies of interest to soccer practitioners. Our approach opens a novel perspective on injury prevention, providing a set of simple and practical rules for evaluating and interpreting the complex relations between injury risk and training performance in professional soccer." }, "1705.09575.pdf": { "id": "1705.09575v3", "title": "Ranking soccer teams on basis of their current strength: a comparison of maximum likelihood approaches", "year": "2017", "authors": [ "Christophe Ley", "Tom Van de Wiele", "Hans Van Eetvelde" ], "categories": "stat.AP", "pdf_url": "https://arxiv.org/pdf/1705.09575v3", "abstract": "We present ten different strength-based statistical models that we use to model soccer match outcomes with the aim of producing a new ranking. The models are of four main types: Thurstone-Mosteller, Bradley-Terry, Independent Poisson and Bivariate Poisson, and their common aspect is that the parameters are estimated via weighted maximum likelihood, the weights being a match importance factor and a time depreciation factor giving less weight to matches that are played a long time ago. Since our goal is to build a ranking reflecting the teams' current strengths, we compare the 10 models on basis of their predictive performance via the Rank Probability Score at the level of both domestic leagues and national teams. We find that the best models are the Bivariate and Independent Poisson models. We then illustrate the versatility and usefulness of our new rankings by means of three examples where the existing rankings fail to provide enough information or lead to peculiar results." }, "1706.02447.pdf": { "id": "1706.02447v1", "title": "Luck is Hard to Beat: The Difficulty of Sports Prediction", "year": "2017", "authors": [ "Raquel YS Aoki", "Renato M Assuncao", "Pedro OS Vaz de Melo" ], "categories": "cs.LG, stat.AP", "pdf_url": "https://arxiv.org/pdf/1706.02447v1", "abstract": "Predicting the outcome of sports events is a hard task. We quantify this difficulty with a coefficient that measures the distance between the observed final results of sports leagues and idealized perfectly balanced competitions in terms of skill. This indicates the relative presence of luck and skill. We collected and analyzed all games from 198 sports leagues comprising 1503 seasons from 84 countries of 4 different sports: basketball, soccer, volleyball and handball. We measured the competitiveness by countries and sports. We also identify in each season which teams, if removed from its league, result in a completely random tournament. Surprisingly, not many of them are needed. As another contribution of this paper, we propose a probabilistic graphical model to learn about the teams' skills and to decompose the relative weights of luck and skill in each game. We break down the skill component into factors associated with the teams' characteristics. The model also allows to estimate as 0.36 the probability that an underdog team wins in the NBA league, with a home advantage adding 0.09 to this probability. As shown in the first part of the paper, luck is substantially present even in the most competitive championships, which partially explains why sophisticated and complex feature-based models hardly beat simple models in the task of forecasting sports' outcomes." }, "1706.04336.pdf": { "id": "1706.04336v1", "title": "Predictive modelling of training loads and injury in Australian football", "year": "2017", "authors": [ "David L. Carey", "Kok-Leong Ong", "Rod Whiteley", "Kay M. Crossley", "Justin Crow", "Meg E. Morris" ], "categories": "stat.AP, stat.ML", "pdf_url": "https://arxiv.org/pdf/1706.04336v1", "abstract": "To investigate whether training load monitoring data could be used to predict injuries in elite Australian football players, data were collected from elite athletes over 3 seasons at an Australian football club. Loads were quantified using GPS devices, accelerometers and player perceived exertion ratings. Absolute and relative training load metrics were calculated for each player each day (rolling average, exponentially weighted moving average, acute:chronic workload ratio, monotony and strain). Injury prediction models (regularised logistic regression, generalised estimating equations, random forests and support vector machines) were built for non-contact, non-contact time-loss and hamstring specific injuries using the first two seasons of data. Injury predictions were generated for the third season and evaluated using the area under the receiver operator characteristic (AUC). Predictive performance was only marginally better than chance for models of non-contact and non-contact time-loss injuries (AUC$<$0.65). The best performing model was a multivariate logistic regression for hamstring injuries (best AUC=0.76). Learning curves suggested logistic regression was underfitting the load-injury relationship and that using a more complex model or increasing the amount of model building data may lead to future improvements. Injury prediction models built using training load data from a single club showed poor ability to predict injuries when tested on previously unseen data, suggesting they are limited as a daily decision tool for practitioners. Focusing the modelling approach on specific injury types and increasing the amount of training data may lead to the development of improved predictive models for injury prevention." }, "1706.04943.pdf": { "id": "1706.04943v1", "title": "Plus-Minus Player Ratings for Soccer", "year": "2017", "authors": [ "Tarak Kharrat", "Javier L\u00f3pez Pe\u00f1a", "Ian McHale" ], "categories": "stat.AP", "pdf_url": "https://arxiv.org/pdf/1706.04943v1", "abstract": "The paper presents a plus-minus rating for use in association football (soccer). We first describe the standard plus-minus methodology as used in basketball and ice-hockey and then adapt it for use in soccer. The usual goal-differential plus-minus is considered before two variations are proposed. For the first variation, we present a methodology to calculate an expected goals plus-minus rating. The second variation makes use of in-play probabilities of match outcome to evaluate an expected points plus-minus rating. We use the ratings to examine who are the best players in European football, and demonstrate how the players' ratings evolve over time. Finally, we shed light on the debate regarding which is the strongest league. The model suggests the English Premier League is the strongest, with the German Bundesliga a close runner-up." }, "1706.06122.pdf": { "id": "1706.06122v2", "title": "VAIN: Attentional Multi-agent Predictive Modeling", "year": "2017", "authors": [ "Yedid Hoshen" ], "categories": "cs.LG, cs.AI", "pdf_url": "https://arxiv.org/pdf/1706.06122v2", "abstract": "Multi-agent predictive modeling is an essential step for understanding physical, social and team-play systems. Recently, Interaction Networks (INs) were proposed for the task of modeling multi-agent physical systems, INs scale with the number of interactions in the system (typically quadratic or higher order in the number of agents). In this paper we introduce VAIN, a novel attentional architecture for multi-agent predictive modeling that scales linearly with the number of agents. We show that VAIN is effective for multi-agent predictive modeling. Our method is evaluated on tasks from challenging multi-agent prediction domains: chess and soccer, and outperforms competing multi-agent approaches." }, "1706.09193.pdf": { "id": "1706.09193v1", "title": "A Parameterized Approach to Personalized Variable Length Summarization of Soccer Matches", "year": "2017", "authors": [ "Mohak Sukhwani", "Ravi Kothari" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/1706.09193v1", "abstract": "We present a parameterized approach to produce personalized variable length summaries of soccer matches. Our approach is based on temporally segmenting the soccer video into 'plays', associating a user-specifiable 'utility' for each type of play and using 'bin-packing' to select a subset of the plays that add up to the desired length while maximizing the overall utility (volume in bin-packing terms). Our approach systematically allows a user to override the default weights assigned to each type of play with individual preferences and thus see a highly personalized variable length summarization of soccer matches. We demonstrate our approach based on the output of an end-to-end pipeline that we are building to produce such summaries. Though aspects of the overall end-to-end pipeline are human assisted at present, the results clearly show that the proposed approach is capable of producing semantically meaningful and compelling summaries. Besides the obvious use of producing summaries of superior league matches for news broadcasts, we anticipate our work to promote greater awareness of the local matches and junior leagues by producing consumable summaries of them." }, "1707.05481.pdf": { "id": "1707.05481v2", "title": "A Comparative Analysis of Social Network Pages by Interests of Their Followers", "year": "2017", "authors": [ "Elena Mikhalkova", "Nadezhda Ganzherli", "Yuri Karyakin" ], "categories": "cs.CL, cs.IR, cs.SI", "pdf_url": "https://arxiv.org/pdf/1707.05481v2", "abstract": "Being a matter of cognition, user interests should be apt to classification independent of the language of users, social network and content of interest itself. To prove it, we analyze a collection of English and Russian Twitter and Vkontakte community pages by interests of their followers. First, we create a model of Major Interests (MaIs) with the help of expert analysis and then classify a set of pages using machine learning algorithms (SVM, Neural Network, Naive Bayes, and some other). We take three interest domains that are typical of both English and Russian-speaking communities: football, rock music, vegetarianism. The results of classification show a greater correlation between Russian-Vkontakte and Russian-Twitter pages while English-Twitterpages appear to provide the highest score." }, "1708.05821.pdf": { "id": "1708.05821v1", "title": "Analysing Soccer Games with Clustering and Conceptors", "year": "2017", "authors": [ "Olivia Michael", "Oliver Obst", "Falk Schmidsberger", "Frieder Stolzenburg" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/1708.05821v1", "abstract": "We present a new approach for identifying situations and behaviours, which we call \"moves\", from soccer games in the 2D simulation league. Being able to identify key situations and behaviours are useful capabilities for analysing soccer matches, anticipating opponent behaviours to aid selection of appropriate tactics, and also as a prerequisite for automatic learning of behaviours and policies. To support a wide set of strategies, our goal is to identify situations from data, in an unsupervised way without making use of pre-defined soccer specific concepts such as \"pass\" or \"dribble\". The recurrent neural networks we use in our approach act as a high-dimensional projection of the recent history of a situation on the field. Similar situations, i.e., with similar histories, are found by clustering of network states. The same networks are also used to learn so-called conceptors, that are lower-dimensional manifolds that describe trajectories through a high-dimensional state space that enable situation-specific predictions from the same neural network. With the proposed approach, we can segment games into sequences of situations that are learnt in an unsupervised way, and learn conceptors that are useful for the prediction of the near future of the respective situation." }, "1711.01703.pdf": { "id": "1711.01703v1", "title": "RoboCupSimData: A RoboCup soccer research dataset", "year": "2017", "authors": [ "Olivia Michael", "Oliver Obst", "Falk Schmidsberger", "Frieder Stolzenburg" ], "categories": "cs.AI, cs.LG, cs.RO", "pdf_url": "https://arxiv.org/pdf/1711.01703v1", "abstract": "RoboCup is an international scientific robot competition in which teams of multiple robots compete against each other. Its different leagues provide many sources of robotics data, that can be used for further analysis and application of machine learning. This paper describes a large dataset from games of some of the top teams (from 2016 and 2017) in RoboCup Soccer Simulation League (2D), where teams of 11 robots (agents) compete against each other. Overall, we used 10 different teams to play each other, resulting in 45 unique pairings. For each pairing, we ran 25 matches (of 10mins), leading to 1125 matches or more than 180 hours of game play. The generated CSV files are 17GB of data (zipped), or 229GB (unzipped). The dataset is unique in the sense that it contains both the ground truth data (global, complete, noise-free information of all objects on the field), as well as the noisy, local and incomplete percepts of each robot. These data are made available as CSV files, as well as in the original soccer simulator formats." }, "1711.05865.pdf": { "id": "1711.05865v2", "title": "Pricing Football Players using Neural Networks", "year": "2017", "authors": [ "Sourya Dey" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/1711.05865v2", "abstract": "We designed a multilayer perceptron neural network to predict the price of a football (soccer) player using data on more than 15,000 players from the football simulation video game FIFA 2017. The network was optimized by experimenting with different activation functions, number of neurons and layers, learning rate and its decay, Nesterov momentum based stochastic gradient descent, L2 regularization, and early stopping. Simultaneous exploration of various aspects of neural network training is performed and their trade-offs are investigated. Our final model achieves a top-5 accuracy of 87.2% among 119 pricing categories, and places any footballer within 6.32% of his actual price on average." }, "1712.07893.pdf": { "id": "1712.07893v2", "title": "A Deep Policy Inference Q-Network for Multi-Agent Systems", "year": "2017", "authors": [ "Zhang-Wei Hong", "Shih-Yang Su", "Tzu-Yun Shann", "Yi-Hsiang Chang", "Chun-Yi Lee" ], "categories": "cs.AI", "pdf_url": "https://arxiv.org/pdf/1712.07893v2", "abstract": "We present DPIQN, a deep policy inference Q-network that targets multi-agent systems composed of controllable agents, collaborators, and opponents that interact with each other. We focus on one challenging issue in such systems---modeling agents with varying strategies---and propose to employ \"policy features\" learned from raw observations (e.g., raw images) of collaborators and opponents by inferring their policies. DPIQN incorporates the learned policy features as a hidden vector into its own deep Q-network (DQN), such that it is able to predict better Q values for the controllable agents than the state-of-the-art deep reinforcement learning models. We further propose an enhanced version of DPIQN, called deep recurrent policy inference Q-network (DRPIQN), for handling partial observability. Both DPIQN and DRPIQN are trained by an adaptive training procedure, which adjusts the network's attention to learn the policy features and its own Q-values at different phases of the training process. We present a comprehensive analysis of DPIQN and DRPIQN, and highlight their effectiveness and generalizability in various multi-agent settings. Our models are evaluated in a classic soccer game involving both competitive and collaborative scenarios. Experimental results performed on 1 vs. 1 and 2 vs. 2 games show that DPIQN and DRPIQN demonstrate superior performance to the baseline DQN and deep recurrent Q-network (DRQN) models. We also explore scenarios in which collaborators or opponents dynamically change their policies, and show that DPIQN and DRPIQN do lead to better overall performance in terms of stability and mean scores." }, "1801.09005.pdf": { "id": "1801.09005v1", "title": "A Two-point Method for PTZ Camera Calibration in Sports", "year": "2018", "authors": [ "Jianhui Chen", "Fangrui Zhu", "James J. Little" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/1801.09005v1", "abstract": "Calibrating narrow field of view soccer cameras is challenging because there are very few field markings in the image. Unlike previous solutions, we propose a two-point method, which requires only two point correspondences given the prior knowledge of base location and orientation of a pan-tilt-zoom (PTZ) camera. We deploy this new calibration method to annotate pan-tilt-zoom data from soccer videos. The collected data are used as references for new images. We also propose a fast random forest method to predict pan-tilt angles without image-to-image feature matching, leading to an efficient calibration method for new images. We demonstrate our system on synthetic data and two real soccer datasets. Our two-point approach achieves superior performance over the state-of-the-art method." }, "1802.00002.pdf": { "id": "1802.00002v1", "title": "DxNAT - Deep Neural Networks for Explaining Non-Recurring Traffic Congestion", "year": "2018", "authors": [ "Fangzhou sun", "Abhishek Dubey", "Jules White" ], "categories": "cs.LG, stat.ML", "pdf_url": "https://arxiv.org/pdf/1802.00002v1", "abstract": "Non-recurring traffic congestion is caused by temporary disruptions, such as accidents, sports games, adverse weather, etc. We use data related to real-time traffic speed, jam factors (a traffic congestion indicator), and events collected over a year from Nashville, TN to train a multi-layered deep neural network. The traffic dataset contains over 900 million data records. The network is thereafter used to classify the real-time data and identify anomalous operations. Compared with traditional approaches of using statistical or machine learning techniques, our model reaches an accuracy of 98.73 percent when identifying traffic congestion caused by football games. Our approach first encodes the traffic across a region as a scaled image. After that the image data from different timestamps is fused with event- and time-related data. Then a crossover operator is used as a data augmentation method to generate training datasets with more balanced classes. Finally, we use the receiver operating characteristic (ROC) analysis to tune the sensitivity of the classifier. We present the analysis of the training time and the inference time separately." }, "1802.03308.pdf": { "id": "1802.03308v10", "title": "Efficient Time-Series Approximation with Linear Recurrent Neural Networks: Architecture Learning and Predictive Power", "year": "2018", "authors": [ "Frieder Stolzenburg", "Sandra Litz", "Olivia Michael", "Oliver Obst" ], "categories": "cs.LG, cs.NE", "pdf_url": "https://arxiv.org/pdf/1802.03308v10", "abstract": "Recurrent neural networks are a powerful means to cope with time series. We show how autoregressive linear, i.e., linearly activated recurrent neural networks (LRNNs) can approximate any time-dependent function f(t). The approximation can effectively be learned by simply solving a linear equation system; no backpropagation or similar methods are needed. Furthermore, and this is the main contribution of this paper, the size of an LRNN can be reduced significantly in one step after inspecting the spectrum of the network transition matrix, i.e., its eigenvalues, by taking only the most relevant components. Therefore, in contrast to other approaches, we do not only learn network weights but also the network architecture. LRNNs have interesting properties: They end up in ellipse trajectories in the long run and allow the prediction of further values and compact representations of functions. We demonstrate this by several case studies, among them multiple superimposed oscillators (MSO), robotic soccer (RoboCup), and stock price prediction. LRNNs outperform the previous state-of-the-art for the MSO task with a minimal number of units." }, "1802.04987.pdf": { "id": "1802.04987v3", "title": "PlayeRank: data-driven performance evaluation and player ranking in soccer via a machine learning approach", "year": "2018", "authors": [ "Luca Pappalardo", "Paolo Cintia", "Paolo Ferragina", "Emanuele Massucco", "Dino Pedreschi", "Fosca Giannotti" ], "categories": "stat.AP, cs.AI", "pdf_url": "https://arxiv.org/pdf/1802.04987v3", "abstract": "The problem of evaluating the performance of soccer players is attracting the interest of many companies and the scientific community, thanks to the availability of massive data capturing all the events generated during a match (e.g., tackles, passes, shots, etc.). Unfortunately, there is no consolidated and widely accepted metric for measuring performance quality in all of its facets. In this paper, we design and implement PlayeRank, a data-driven framework that offers a principled multi-dimensional and role-aware evaluation of the performance of soccer players. We build our framework by deploying a massive dataset of soccer-logs and consisting of millions of match events pertaining to four seasons of 18 prominent soccer competitions. By comparing PlayeRank to known algorithms for performance evaluation in soccer, and by exploiting a dataset of players' evaluations made by professional soccer scouts, we show that PlayeRank significantly outperforms the competitors. We also explore the ratings produced by {\\sf PlayeRank} and discover interesting patterns about the nature of excellent performances and what distinguishes the top players from the others. At the end, we explore some applications of PlayeRank -- i.e. searching players and player versatility --- showing its flexibility and efficiency, which makes it worth to be used in the design of a scalable platform for soccer analytics." }, "1802.07127.pdf": { "id": "1802.07127v2", "title": "Actions Speak Louder Than Goals: Valuing Player Actions in Soccer", "year": "2018", "authors": [ "Tom Decroos", "Lotte Bransen", "Jan Van Haaren", "Jesse Davis" ], "categories": "stat.AP, stat.ML", "pdf_url": "https://arxiv.org/pdf/1802.07127v2", "abstract": "Assessing the impact of the individual actions performed by soccer players during games is a crucial aspect of the player recruitment process. Unfortunately, most traditional metrics fall short in addressing this task as they either focus on rare actions like shots and goals alone or fail to account for the context in which the actions occurred. This paper introduces (1) a new language for describing individual player actions on the pitch and (2) a framework for valuing any type of player action based on its impact on the game outcome while accounting for the context in which the action happened. By aggregating soccer players' action values, their total offensive and defensive contributions to their team can be quantified. We show how our approach considers relevant contextual information that traditional player evaluation metrics ignore and present a number of use cases related to scouting and playing style characterization in the 2016/2017 and 2017/2018 seasons in Europe's top competitions." }, "1803.07981.pdf": { "id": "1803.07981v1", "title": "A time-dependent scoring system for soccer leagues", "year": "2018", "authors": [ "Manuel Cruz", "Sandra Ramos", "Miguel Pinho" ], "categories": "stat.AP", "pdf_url": "https://arxiv.org/pdf/1803.07981v1", "abstract": "In this paper, a new continuous scoring system for soccer is proposed, based on the proportion of time that a team is winning, losing or tied. Several simulations are made applying this technique to complete seasons of different leagues. As some preliminary fundamental analysis previews, the simulation results show that this proposal compacts the gap between several teams, increasing the competitiveness of the championships and making the matches even more challenging as every minute may count in order to get the best rank in the final standings. Based on those results, some characteristics of the model are highlighted and a discussion on some of the advantages, disadvantages and practical issues regarding a hypothetical implementation of this model is made. Finally, some ideas to tackle the less positive effects of an implementation in real leagues is also provided." }, "1804.04527.pdf": { "id": "1804.04527v2", "title": "SoccerNet: A Scalable Dataset for Action Spotting in Soccer Videos", "year": "2018", "authors": [ "Silvio Giancola", "Mohieddine Amine", "Tarek Dghaily", "Bernard Ghanem" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/1804.04527v2", "abstract": "In this paper, we introduce SoccerNet, a benchmark for action spotting in soccer videos. The dataset is composed of 500 complete soccer games from six main European leagues, covering three seasons from 2014 to 2017 and a total duration of 764 hours. A total of 6,637 temporal annotations are automatically parsed from online match reports at a one minute resolution for three main classes of events (Goal, Yellow/Red Card, and Substitution). As such, the dataset is easily scalable. These annotations are manually refined to a one second resolution by anchoring them at a single timestamp following well-defined soccer rules. With an average of one event every 6.9 minutes, this dataset focuses on the problem of localizing very sparse events within long videos. We define the task of spotting as finding the anchors of soccer events in a video. Making use of recent developments in the realm of generic action recognition and detection in video, we provide strong baselines for detecting soccer events. We show that our best model for classifying temporal segments of length one minute reaches a mean Average Precision (mAP) of 67.8%. For the spotting task, our baseline reaches an Average-mAP of 49.7% for tolerances $\u03b4$ ranging from 5 to 60 seconds. Our dataset and models are available at https://silviogiancola.github.io/SoccerNet." }, "1805.05009.pdf": { "id": "1805.05009v1", "title": "Deep Decision Trees for Discriminative Dictionary Learning with Adversarial Multi-Agent Trajectories", "year": "2018", "authors": [ "Tharindu Fernando", "Sridha Sridharan", "Clinton Fookes", "Simon Denman" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/1805.05009v1", "abstract": "With the explosion in the availability of spatio-temporal tracking data in modern sports, there is an enormous opportunity to better analyse, learn and predict important events in adversarial group environments. In this paper, we propose a deep decision tree architecture for discriminative dictionary learning from adversarial multi-agent trajectories. We first build up a hierarchy for the tree structure by adding each layer and performing feature weight based clustering in the forward pass. We then fine tune the player role weights using back propagation. The hierarchical architecture ensures the interpretability and the integrity of the group representation. The resulting architecture is a decision tree, with leaf-nodes capturing a dictionary of multi-agent group interactions. Due to the ample volume of data available, we focus on soccer tracking data, although our approach can be used in any adversarial multi-agent domain. We present applications of proposed method for simulating soccer games as well as evaluating and quantifying team strategies." }, "1805.10933.pdf": { "id": "1805.10933v2", "title": "Analysis of association football playing styles: an innovative method to cluster networks", "year": "2018", "authors": [ "Jacopo Diquigiovanni", "Bruno Scarpa" ], "categories": "stat.AP", "pdf_url": "https://arxiv.org/pdf/1805.10933v2", "abstract": "In this work we develop an innovative hierarchical clustering method to divide a sample of undirected weighted networks into groups. The methodology consists of two phases: the first phase is aimed at putting the single networks in a broader framework by including the characteristics of the population in the data, while the second phase creates a subdivision of the sample on the basis of the similarity between the community structures of the processed networks. Starting from the representation of the team's playing style as a network, we apply the method to group the Italian Serie A teams' performances and consequently detect the main 15 tactics shown during the 2015-2016 season. The information obtained is used to verify the effect of the styles of play on the number of goals scored, and we prove the key role of one of them by implementing an extension of the Dixon and Coles model (Dixon and Coles, 1997)." }, "1806.00890.pdf": { "id": "1806.00890v1", "title": "Soccer on Your Tabletop", "year": "2018", "authors": [ "Konstantinos Rematas", "Ira Kemelmacher-Shlizerman", "Brian Curless", "Steve Seitz" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/1806.00890v1", "abstract": "We present a system that transforms a monocular video of a soccer game into a moving 3D reconstruction, in which the players and field can be rendered interactively with a 3D viewer or through an Augmented Reality device. At the heart of our paper is an approach to estimate the depth map of each player, using a CNN that is trained on 3D player data extracted from soccer video games. We compare with state of the art body pose and depth estimation techniques, and show results on both synthetic ground truth benchmarks, and real YouTube soccer footage." }, "1806.01114.pdf": { "id": "1806.01114v11", "title": "A comparison of penalty shootout designs in soccer", "year": "2018", "authors": [ "L\u00e1szl\u00f3 Csat\u00f3" ], "categories": "cs.GT", "pdf_url": "https://arxiv.org/pdf/1806.01114v11", "abstract": "Penalty shootout in soccer is recognized to be unfair because the team kicking first in all rounds enjoys a significant advantage. The so-called Catch-Up Rule has been suggested recently to solve this problem but is shown here not to be fairer than the simpler deterministic Alternating (ABBA) Rule that has already been tried. We introduce the Adjusted Catch-Up Rule by guaranteeing the first penalty of the possible sudden death stage to the team disadvantaged in the first round. It outperforms the Catch-Up and Alternating Rules, while remains straightforward to implement. A general measure of complexity for penalty shootout mechanisms is also provided as the minimal number of binary questions required to decide the first-mover in a given round without knowing the history of the penalty shootout. This quantification permits a two-dimensional evaluation of any mechanism proposed in the future." }, "1806.01595.pdf": { "id": "1806.01595v1", "title": "A penalty criterion for score forecasting in soccer", "year": "2018", "authors": [ "Jean-Louis Foulley", "Gilles Celeux" ], "categories": "stat.AP", "pdf_url": "https://arxiv.org/pdf/1806.01595v1", "abstract": "This note proposes a penalty criterion for assessing correct score forecasting in a soccer match. The penalty is based on hierarchical priorities for such a forecast i.e., i) Win, Draw and Loss exact prediction and ii) normalized Euclidian distance between actual and forecast scores. The procedure is illustrated on typical scores, and different alternatives on the penalty components are discussed." }, "1806.03208.pdf": { "id": "1806.03208v3", "title": "Prediction of the FIFA World Cup 2018 - A random forest approach with an emphasis on estimated team ability parameters", "year": "2018", "authors": [ "Andreas Groll", "Christophe Ley", "Gunther Schauberger", "Hans Van Eetvelde" ], "categories": "stat.AP", "pdf_url": "https://arxiv.org/pdf/1806.03208v3", "abstract": "In this work, we compare three different modeling approaches for the scores of soccer matches with regard to their predictive performances based on all matches from the four previous FIFA World Cups 2002 - 2014: Poisson regression models, random forests and ranking methods. While the former two are based on the teams' covariate information, the latter method estimates adequate ability parameters that reflect the current strength of the teams best. Within this comparison the best-performing prediction methods on the training data turn out to be the ranking methods and the random forests. However, we show that by combining the random forest with the team ability parameters from the ranking methods as an additional covariate we can improve the predictive power substantially. Finally, this combination of methods is chosen as the final model and based on its estimates, the FIFA World Cup 2018 is simulated repeatedly and winning probabilities are obtained for all teams. The model slightly favors Spain before the defending champion Germany. Additionally, we provide survival probabilities for all teams and at all tournament stages as well as the most probable tournament outcome." }, "1807.01623.pdf": { "id": "1807.01623v2", "title": "Modeling outcomes of soccer matches", "year": "2018", "authors": [ "Alkeos Tsokos", "Santhosh Narayanan", "Ioannis Kosmidis", "Gianluca Baio", "Mihai Cucuringu", "Gavin Whitaker" ], "categories": "stat.AP, stat.ML", "pdf_url": "https://arxiv.org/pdf/1807.01623v2", "abstract": "We compare various extensions of the Bradley-Terry model and a hierarchical Poisson log-linear model in terms of their performance in predicting the outcome of soccer matches (win, draw, or loss). The parameters of the Bradley-Terry extensions are estimated by maximizing the log-likelihood, or an appropriately penalized version of it, while the posterior densities of the parameters of the hierarchical Poisson log-linear model are approximated using integrated nested Laplace approximations. The prediction performance of the various modeling approaches is assessed using a novel, context-specific framework for temporal validation that is found to deliver accurate estimates of the test error. The direct modeling of outcomes via the various Bradley-Terry extensions and the modeling of match scores using the hierarchical Poisson log-linear model demonstrate similar behavior in terms of predictive performance." }, "1807.05059.pdf": { "id": "1807.05059v1", "title": "Cyber-Physical System for Energy-Efficient Stadium Operation: Methodology and Experimental Validation", "year": "2018", "authors": [ "Mischa Schmidt", "Anett Sch\u00fclke", "Alberto Venturi", "Roman Kurpatov", "Enrique Blanco Henriquez" ], "categories": "eess.SY", "pdf_url": "https://arxiv.org/pdf/1807.05059v1", "abstract": "The environmental impacts of medium to large scale buildings receive substantial attention in research, industry, and media. This paper studies the energy savings potential of a commercial soccer stadium during day-to-day operation. Buildings of this kind are characterized by special purpose system installations like grass heating systems and by event-driven usage patterns. This work presents a methodology to holistically analyze the stadiums characteristics and integrate its existing instrumentation into a Cyber-Physical System, enabling to deploy different control strategies flexibly. In total, seven different strategies for controlling the studied stadiums grass heating system are developed and tested in operation. Experiments in winter season 2014/2015 validated the strategies impacts within the real operational setup of the Commerzbank Arena, Frankfurt, Germany. With 95% confidence, these experiments saved up to 66% of median daily weather-normalized energy consumption. Extrapolated to an average heating season, this corresponds to savings of 775 MWh and 148 t of CO2 emissions. In winter 2015/2016 an additional predictive nighttime heating experiment targeted lower temperatures, which increased the savings to up to 85%, equivalent to 1 GWh (197 t CO2) in an average winter. Beyond achieving significant energy savings, the different control strategies also met the target temperature levels to the satisfaction of the stadiums operational staff. While the case study constitutes a significant part, the discussions dedicated to the transferability of this work to other stadiums and other building types show that the concepts and the approach are of general nature. Furthermore, this work demonstrates the first successful application of Deep Belief Networks to regress and predict the thermal evolution of building systems." }, "1807.05597.pdf": { "id": "1807.05597v1", "title": "Deep Learning for Semantic Segmentation on Minimal Hardware", "year": "2018", "authors": [ "Sander G. van Dijk", "Marcus M. Scheunemann" ], "categories": "cs.LG, cs.CV, cs.RO, stat.ML", "pdf_url": "https://arxiv.org/pdf/1807.05597v1", "abstract": "Deep learning has revolutionised many fields, but it is still challenging to transfer its success to small mobile robots with minimal hardware. Specifically, some work has been done to this effect in the RoboCup humanoid football domain, but results that are performant and efficient and still generally applicable outside of this domain are lacking. We propose an approach conceptually different from those taken previously. It is based on semantic segmentation and does achieve these desired properties. In detail, it is being able to process full VGA images in real-time on a low-power mobile processor. It can further handle multiple image dimensions without retraining, it does not require specific domain knowledge for achieving a high frame rate and it is applicable on a minimal mobile hardware." }, "1807.07536.pdf": { "id": "1807.07536v5", "title": "A Skellam Regression Model for Quantifying Positional Value in Soccer", "year": "2018", "authors": [ "Konstantinos Pelechrinis", "Wayne Winston" ], "categories": "stat.AP", "pdf_url": "https://arxiv.org/pdf/1807.07536v5", "abstract": "Soccer is undeniably the most popular sport world-wide and everyone from general managers and coaching staff to fans and media are interested in evaluating players' performance. Metrics applied successfully in other sports, such as the (adjusted) +/- that allows for division of credit among a basketball team's players, exhibit several challenges when applied to soccer due to severe co-linearities. Recently, a number of player evaluation metrics have been developed utilizing optical tracking data, but they are based on proprietary data. In this work, our objective is to develop an open framework that can estimate the expected contribution of a soccer player to his team's winning chances using publicly available data. In particular, using data from (i) approximately 20,000 games from 11 European leagues over 8 seasons, and, (ii) player ratings from the FIFA video game, we estimate through a Skellam regression model the importance of every line (attackers, midfielders, defenders and goalkeeping) in winning a soccer game. We consequently translate the model to expected league points added above a replacement player (eLPAR). This model can further be used as a guide for allocating a team's salary budget to players based on their expected contributions on the pitch. We showcase similar applications using annual salary data from the English Premier League and identify evidence that in our dataset the market appears to under-value defensive line players relative to goalkeepers." }, "1808.04550.pdf": { "id": "1808.04550v1", "title": "SciSports: Learning football kinematics through two-dimensional tracking data", "year": "2018", "authors": [ "Anatoliy Babic", "Harshit Bansal", "Gianluca Finocchio", "Julian Golak", "Mark Peletier", "Jim Portegies" ], "categories": "cs.LG, stat.ML", "pdf_url": "https://arxiv.org/pdf/1808.04550v1", "abstract": "SciSports is a Dutch startup company specializing in football analytics. This paper describes a joint research effort with SciSports, during the Study Group Mathematics with Industry 2018 at Eindhoven, the Netherlands. The main challenge that we addressed was to automatically process empirical football players' trajectories, in order to extract useful information from them. The data provided to us was two-dimensional positional data during entire matches. We developed methods based on Newtonian mechanics and the Kalman filter, Generative Adversarial Nets and Variational Autoencoders. In addition, we trained a discriminator network to recognize and discern different movement patterns of players. The Kalman-filter approach yields an interpretable model, in which a small number of player-dependent parameters can be fit; in theory this could be used to distinguish among players. The Generative-Adversarial-Nets approach appears promising in theory, and some initial tests showed an improvement with respect to the baseline, but the limits in time and computational power meant that we could not fully explore it. We also trained a Discriminator network to distinguish between two players based on their trajectories; after training, the network managed to distinguish between some pairs of players, but not between others. After training, the Variational Autoencoders generated trajectories that are difficult to distinguish, visually, from the data. These experiments provide an indication that deep generative models can learn the underlying structure and statistics of football players' trajectories. This can serve as a starting point for determining player qualities based on such trajectory data." }, "1809.02854.pdf": { "id": "1809.02854v1", "title": "Learning Sports Camera Selection from Internet Videos", "year": "2018", "authors": [ "Jianhui Chen", "Keyu Lu", "Sijia Tian", "James J. Little" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/1809.02854v1", "abstract": "This work addresses camera selection, the task of predicting which camera should be \"on air\" from multiple candidate cameras for soccer broadcast. The task is challenging because of the scarcity of learning data with all candidate views. Meanwhile, broadcast videos are freely available on the Internet (e.g. Youtube). However, these videos only record the selected camera views, omitting the other candidate views. To overcome this problem, we first introduce a random survival forest (RSF) method to impute the incomplete data effectively. Then, we propose a spatial-appearance heatmap to describe foreground objects (e.g. players and balls) in an image. To evaluate the performance of our system, we collect the largest-ever dataset for soccer broadcasting camera selection. It has one main game which has all candidate views and twelve auxiliary games which only have the broadcast view. Our method significantly outperforms state-of-the-art methods on this challenging dataset. Further analysis suggests that the improvement in performance is indeed from the extra information from auxiliary games." }, "1809.04928.pdf": { "id": "1809.04928v1", "title": "Grown-up NimbRo Robots Winning RoboCup 2017 Humanoid AdultSize Soccer Competitions", "year": "2018", "authors": [ "Grzegorz Ficht", "Dmytro Pavlichenko", "Philipp Allgeuer", "Hafez Farazi", "Diego Rodriguez", "Andr\u00e9 Brandenburger" ], "categories": "cs.RO", "pdf_url": "https://arxiv.org/pdf/1809.04928v1", "abstract": "The ongoing evolution of the RoboCup Humanoid League led in 2017 to the introduction of one vs. one soccer games for the AdultSize robots, which motived our team NimbRo to enter this category. In this paper, we present the mechatronic design of our upgraded robot Copedo and the newly developed NimbRo-OP2, which received the RoboCup Design Award. We also describe improved approaches to visual perception of the game situation, including compassless localization on a soccer field with symmetric appearance, and the generation of soccer behaviors. At RoboCup 2017 in Nagoya, our robots played very well, winning the AdultSize soccer tournament with high scores. Our robots also won the technical challenges and we present the developed solutions." }, "1809.05173.pdf": { "id": "1809.05173v1", "title": "Distinguishing Between Roles of Football Players in Play-by-play Match Event Data", "year": "2018", "authors": [ "Bart Aalbers", "Jan Van Haaren" ], "categories": "stat.ML, cs.LG, stat.AP", "pdf_url": "https://arxiv.org/pdf/1809.05173v1", "abstract": "Over the last few decades, the player recruitment process in professional football has evolved into a multi-billion industry and has thus become of vital importance. To gain insights into the general level of their candidate reinforcements, many professional football clubs have access to extensive video footage and advanced statistics. However, the question whether a given player would fit the team's playing style often still remains unanswered. In this paper, we aim to bridge that gap by proposing a set of 21 player roles and introducing a method for automatically identifying the most applicable roles for each player from play-by-play event data collected during matches." }, "1809.05361.pdf": { "id": "1809.05361v1", "title": "Advanced Soccer Skills and Team Play of RoboCup 2017 TeenSize Winner NimbRo", "year": "2018", "authors": [ "Diego Rodriguez", "Hafez Farazi", "Philipp Allgeuer", "Dmytro Pavlichenko", "Grzegorz Ficht", "Andr\u00e9 Brandenburger" ], "categories": "cs.RO", "pdf_url": "https://arxiv.org/pdf/1809.05361v1", "abstract": "In order to pursue the vision of the RoboCup Humanoid League of beating the soccer world champion by 2050, new rules and competitions are added or modified each year fostering novel technological advances. In 2017, the number of players in the TeenSize class soccer games was increase to 3 vs. 3, which allowed for more team play strategies. Improvements in individual skills were also demanded through a set of technical challenges. This paper presents the latest individual skills and team play developments used in RoboCup 2017 that lead our team Nimbro winning the 2017 TeenSize soccer tournament, the technical challenges, and the drop-in games." }, "1809.08039.pdf": { "id": "1809.08039v1", "title": "Besov and Triebel-Lizorkin spaces associated with Laguerre expansions of Hermite type", "year": "2018", "authors": [ "Pawe\u0142 Plewa" ], "categories": "math.CA", "pdf_url": "https://arxiv.org/pdf/1809.08039v1", "abstract": "Homogeneous Besov and Triebel-Lizorkin spaces associated with multi-dimensional Laguerre function expansions of Hermite type with index $\u03b1\\in [-1/2,\\infty)^d\\backslash (-1/2,1/2)^d$, $d\\geq 1$, are defined and investigated. To achieve expected goals Schwartz type spaces on $R^d_+$ are introduced and then tempered type distributions are constructed. Also, ideas from a recent paper of Bui and Duong on Besov and Triebel-Lizorkin spaces associated with Hermite functions expansions are used. This means, in particular, using molecular decomposition and an appropriate form of a Calder\u00f3n reproducing formula." }, "1809.11051.pdf": { "id": "1809.11051v1", "title": "A ROS-based Software Framework for the NimbRo-OP Humanoid Open Platform", "year": "2018", "authors": [ "Philipp Allgeuer", "Max Schwarz", "Julio Pastrana", "Sebastian Schueller", "Marcell Missura", "Sven Behnke" ], "categories": "cs.RO", "pdf_url": "https://arxiv.org/pdf/1809.11051v1", "abstract": "Over the past few years, a number of successful humanoid platforms have been developed, including the Nao and the DARwIn-OP, both of which are used by many research groups for the investigation of bipedal walking, full-body motions, and human-robot interaction. The NimbRo-OP is an open humanoid platform under development by team NimbRo of the University of Bonn. Significantly larger than the two aforementioned humanoids, this platform has the potential to interact with a more human-scale environment. This paper describes a software framework for the NimbRo-OP that is based on the Robot Operating System (ROS) middleware. The software provides functionality for hardware abstraction, visual perception, and behavior generation, and has been used to implement basic soccer skills. These were demonstrated at RoboCup 2013, as part of the winning team of the Humanoid League competition." }, "1809.11067.pdf": { "id": "1809.11067v1", "title": "Hierarchical and State-based Architectures for Robot Behavior Planning and Control", "year": "2018", "authors": [ "Philipp Allgeuer", "Sven Behnke" ], "categories": "cs.RO", "pdf_url": "https://arxiv.org/pdf/1809.11067v1", "abstract": "In this paper, two behavior control architectures for autonomous agents in the form of cross-platform C++ frameworks are presented, the State Controller Library and the Behavior Control Framework. While the former is state-based and generalizes the notion of states and finite state machines to allow for multi-action planning, the latter is behavior-based and exploits a hierarchical structure and the concept of inhibitions to allow for dynamic transitioning. The two frameworks have completely independent implementations, but can be used effectively in tandem to solve behavior control problems on all levels of granularity. Both frameworks have been used to control the NimbRo-OP, a humanoid soccer robot developed by team NimbRo of the University of Bonn." }, "1809.11078.pdf": { "id": "1809.11078v1", "title": "A Monocular Vision System for Playing Soccer in Low Color Information Environments", "year": "2018", "authors": [ "Hafez Farazi", "Philipp Allgeuer", "Sven Behnke" ], "categories": "cs.RO", "pdf_url": "https://arxiv.org/pdf/1809.11078v1", "abstract": "Humanoid soccer robots perceive their environment exclusively through cameras. This paper presents a monocular vision system that was originally developed for use in the RoboCup Humanoid League, but is expected to be transferable to other soccer leagues. Recent changes in the Humanoid League rules resulted in a soccer environment with less color coding than in previous years, which makes perception of the game situation more challenging. The proposed vision system addresses these challenges by using brightness and texture for the detection of the required field features and objects. Our system is robust to changes in lighting conditions, and is designed for real-time use on a humanoid soccer robot. This paper describes the main components of the detection algorithms in use, and presents experimental results from the soccer field, using ROS and the igus Humanoid Open Platform as a testbed. The proposed vision system was used successfully at RoboCup 2015." }, "1810.02252.pdf": { "id": "1810.02252v1", "title": "Measuring Football Players' On-the-ball Contributions From Passes During Games", "year": "2018", "authors": [ "Lotte Bransen", "Jan Van Haaren" ], "categories": "stat.AP, cs.LG", "pdf_url": "https://arxiv.org/pdf/1810.02252v1", "abstract": "Several performance metrics for quantifying the in-game performances of individual football players have been proposed in recent years. Although the majority of the on-the-ball actions during games constitutes of passes, many of the currently available metrics focus on measuring the quality of shots only. To help bridge this gap, we propose a novel approach to measure players' on-the-ball contributions from passes during games. Our proposed approach measures the expected impact of each pass on the scoreline." }, "1810.07550.pdf": { "id": "1810.07550v1", "title": "The Newton Scheme for Deep Learning", "year": "2018", "authors": [ "Junqing Qiu", "Guoren Zhong", "Yihua Lu", "Kun Xin", "Huihuan Qian", "Xi Zhu" ], "categories": "cs.LG, physics.class-ph", "pdf_url": "https://arxiv.org/pdf/1810.07550v1", "abstract": "We introduce a neural network (NN) strictly governed by Newton's Law, with the nature required basis functions derived from the fundamental classic mechanics. Then, by classifying the training model as a quick procedure of 'force pattern' recognition, we developed the Newton physics-based NS scheme. Once the force pattern is confirmed, the neuro network simply does the checking of the 'pattern stability' instead of the continuous fitting by computational resource consuming big data-driven processing. In the given physics's law system, once the field is confirmed, the mathematics bases for the force field description actually are not diverged but denumerable, which can save the function representations from the exhaustible available mathematics bases. In this work, we endorsed Newton's Law into the deep learning technology and proposed Newton Scheme (NS). Under NS, the user first identifies the path pattern, like the constant acceleration movement.The object recognition technology first loads mass information, then, the NS finds the matched physical pattern and describe and predict the trajectory of the movements with nearly zero error. We compare the major contribution of this NS with the TCN, GRU and other physics inspired 'FIND-PDE' methods to demonstrate fundamental and extended applications of how the NS works for the free-falling, pendulum and curve soccer balls.The NS methodology provides more opportunity for the future deep learning advances." }, "1810.08032.pdf": { "id": "1810.08032v1", "title": "Augmenting Adjusted Plus-Minus in Soccer with FIFA Ratings", "year": "2018", "authors": [ "Francesca Matano", "Lee F. Richardson", "Taylor Pospisil", "Collin Eubanks", "Jining Qin" ], "categories": "stat.AP", "pdf_url": "https://arxiv.org/pdf/1810.08032v1", "abstract": "In basketball and hockey, state-of-the-art player value statistics are often variants of Adjusted Plus-Minus (APM). But APM hasn't had the same impact in soccer, since soccer games are low scoring with a low number of substitutions. In soccer, perhaps the most comprehensive player value statistics come from video games, and in particular FIFA. FIFA ratings combine the subjective evaluations of over 9000 scouts, coaches, and season-ticket holders into ratings for over 18,000 players. This paper combines FIFA ratings and APM into a single metric, which we call Augmented APM. The key idea is recasting APM into a Bayesian framework, and incorporating FIFA ratings into the prior distribution. We show that Augmented APM predicts better than both standard APM and a model using only FIFA ratings. We also show that Augmented APM decorrelates players that are highly collinear." }, "1810.11593.pdf": { "id": "1810.11593v1", "title": "Reagent: Converting Ordinary Webpages into Interactive Software Agents", "year": "2018", "authors": [ "Mathew Peveler", "Jeffery Kephart", "Hui Su" ], "categories": "cs.HC", "pdf_url": "https://arxiv.org/pdf/1810.11593v1", "abstract": "We introduce Reagent, a technology that readily converts ordinary webpages containing structured data into software agents with which one can interact naturally, via a combination of speech and pointing. Previous efforts to make webpage content manipulable by third-party software components in browsers or desktop applications have generally relied upon specialized instrumentation included in the webpages -- a practice that neither scales well nor applies to pre-existing webpages. In contrast, Reagent automatically captures semantic details and semantically-meaningful mouse events from arbitrary webpages that contain no pre-existing special instrumentation. Reagent combines these events with text transcriptions of user speech to derive and execute parameterized commands representing human intent. Thus, users may request various visualization or analytic operations to be performed on data displayed on a page by speaking to it and/or pointing to elements within it. When unable to infer translations between event labels and human terminology, Reagent proactively asks users for definitions and adds them to its dictionary. We demonstrate Reagent in the context of a collection of pre-existing webpages that contain football team and player statistics." }, "1812.10202.pdf": { "id": "1812.10202v1", "title": "Gliders2d: Source Code Base for RoboCup 2D Soccer Simulation League", "year": "2018", "authors": [ "Mikhail Prokopenko", "Peter Wang" ], "categories": "cs.MA", "pdf_url": "https://arxiv.org/pdf/1812.10202v1", "abstract": "We describe Gliders2d, a base code release for Gliders, a soccer simulation team which won the RoboCup Soccer 2D Simulation League in 2016. We trace six evolutionary steps, each of which is encapsulated in a sequential change of the released code, from v1.1 to v1.6, starting from agent2d-3.1.1 (set as the baseline v1.0). These changes improve performance by adjusting the agents' stamina management, their pressing behaviour and the action-selection mechanism, as well as their positional choice in both attack and defense, and enabling riskier passes. The resultant behaviour, which is sufficiently generic to be applicable to physical robot teams, increases the players' mobility and achieves a better control of the field. The last presented version, Gliders2d-v1.6, approaches the strength of Gliders2013, and outperforms agent2d-3.1.1 by four goals per game on average. The sequential improvements demonstrate how the methodology of human-based evolutionary computation can markedly boost the overall performance with even a small number of controlled steps." }, "1902.09641.pdf": { "id": "1902.09641v1", "title": "Stochastic Prediction of Multi-Agent Interactions from Partial Observations", "year": "2019", "authors": [ "Chen Sun", "Per Karlsson", "Jiajun Wu", "Joshua B Tenenbaum", "Kevin Murphy" ], "categories": "cs.LG, cs.CV, stat.ML", "pdf_url": "https://arxiv.org/pdf/1902.09641v1", "abstract": "We present a method that learns to integrate temporal information, from a learned dynamics model, with ambiguous visual information, from a learned vision model, in the context of interacting agents. Our method is based on a graph-structured variational recurrent neural network (Graph-VRNN), which is trained end-to-end to infer the current state of the (partially observed) world, as well as to forecast future states. We show that our method outperforms various baselines on two sports datasets, one based on real basketball trajectories, and one generated by a soccer game engine." }, "1904.08310.pdf": { "id": "1904.08310v1", "title": "LegenDary 2012 Soccer 2D Simulation Team Description Paper", "year": "2019", "authors": [ "Pourya Saljoughi", "Reza Ma'anijou", "Ehsan Fouladi", "Narges Majidi", "Saber Yaghoobi", "Houman Fallah" ], "categories": "cs.MA, cs.RO", "pdf_url": "https://arxiv.org/pdf/1904.08310v1", "abstract": "In LegenDary project, we started a new research based on Agent2D in RoboCup 2D soccer simulation. In this paper, we mainly present the team algorithms and structures which we used to develop our team in separated section. We have focused on passing, dribbling and blocking skills. We improved them and made the team ready for this competition. Through pass is the most important part of our team that we work a lot on it." }, "1906.02491.pdf": { "id": "1906.02491v1", "title": "Optimized Deployment of Millimeter Wave Networks for In-venue Regions with Stochastic Users' Orientation", "year": "2019", "authors": [ "Mehdi Naderi Soorki", "Walid Saad", "Mehdi Bennis" ], "categories": "cs.NI, cs.IT", "pdf_url": "https://arxiv.org/pdf/1906.02491v1", "abstract": "Millimeter wave (mmW) communication is a promising solution for providing high-capacity wireless network access. However, the benefits of mmW are limited by the fact that the channel between a mmW access point and the user equipment can stochastically change due to severe blockage of mmW links by obstacles such as the human body. Thus, one main challenge of mmW network coverage is to enable directional line-of-sight links between access points and mobile devices. In this paper, a novel framework is proposed for optimizing mmW network coverage within hotspots and in-venue regions, while being cognizant of the body blockage of the network's users. In the studied model, the locations of potential access points and users are assumed as predefined parameters while the orientation of the users is assumed to be stochastic. Hence, a joint stochastic access point placement and beam steering problem subjected to stochastic users' body blockage is formulated, under desired network coverage constraints. Then, a greedy algorithm is introduced to find an approximation solution for the joint deployment and assignment problem using a new `size constrained weighted set cover' approach. A closed-form expression for the ratio between the optimal solution and approximate one (resulting from the greedy algorithm) is analytically derived. The proposed algorithm is simulated for three in-venue regions: the meeting room in the Alumni Assembly Hall of Virginia Tech, an airport gate, and one side of a stadium football. Simulation results show that, in order to guarantee network coverage for different in-venue regions, the greedy algorithm uses at most three more access points (APs) compared to the optimal solution. The results also show that, due to the use of the additional APs, the greedy algorithm will yield a network coverage up to 11.7 % better than the optimal, AP-minimizing solution." }, "1907.01221.pdf": { "id": "1907.01221v1", "title": "Visual analytics for team-based invasion sports with significant events and Markov reward process", "year": "2019", "authors": [ "Kun Zhao", "Takayuki Osogami", "Tetsuro Morimura" ], "categories": "cs.AI, cs.HC, cs.LG", "pdf_url": "https://arxiv.org/pdf/1907.01221v1", "abstract": "In team-based invasion sports such as soccer and basketball, analytics is important for teams to understand their performance and for audiences to understand matches better. The present work focuses on performing visual analytics to evaluate the value of any kind of event occurring in a sports match with a continuous parameter space. Here, the continuous parameter space involves the time, location, score, and other parameters. Because the spatiotemporal data used in such analytics is a low-level representation and has a very large size, however, traditional analytics may need to discretize the continuous parameter space (e.g., subdivide the playing area) or use a local feature to limit the analysis to specific events (e.g., only shots). These approaches make evaluation impossible for any kind of event with a continuous parameter space. To solve this problem, we consider a whole match as a Markov chain of significant events, so that event values can be estimated with a continuous parameter space by solving the Markov chain with a machine learning model. The significant events are first extracted by considering the time-varying distribution of players to represent the whole match. Then, the extracted events are redefined as different states with the continuous parameter space and built as a Markov chain so that a Markov reward process can be applied. Finally, the Markov reward process is solved by a customized fitted-value iteration algorithm so that the event values with the continuous parameter space can be predicted by a regression model. As a result, the event values can be visually inspected over the whole playing field under arbitrary given conditions. Experimental results with real soccer data show the effectiveness of the proposed system." }, "1907.11180.pdf": { "id": "1907.11180v2", "title": "Google Research Football: A Novel Reinforcement Learning Environment", "year": "2019", "authors": [ "Karol Kurach", "Anton Raichuk", "Piotr Sta\u0144czyk", "Micha\u0142 Zaj\u0105c", "Olivier Bachem", "Lasse Espeholt" ], "categories": "cs.LG, stat.ML", "pdf_url": "https://arxiv.org/pdf/1907.11180v2", "abstract": "Recent progress in the field of reinforcement learning has been accelerated by virtual learning environments such as video games, where novel algorithms and ideas can be quickly tested in a safe and reproducible manner. We introduce the Google Research Football Environment, a new reinforcement learning environment where agents are trained to play football in an advanced, physics-based 3D simulator. The resulting environment is challenging, easy to use and customize, and it is available under a permissive open-source license. In addition, it provides support for multiplayer and multi-agent experiments. We propose three full-game scenarios of varying difficulty with the Football Benchmarks and report baseline results for three commonly used reinforcement algorithms (IMPALA, PPO, and Ape-X DQN). We also provide a diverse set of simpler scenarios with the Football Academy and showcase several promising research directions." }, "1908.00698.pdf": { "id": "1908.00698v2", "title": "Soccer Team Vectors", "year": "2019", "authors": [ "Robert M\u00fcller", "Stefan Langer", "Fabian Ritz", "Christoph Roch", "Steffen Illium", "Claudia Linnhoff-Popien" ], "categories": "cs.LG, cs.AI, stat.ML", "pdf_url": "https://arxiv.org/pdf/1908.00698v2", "abstract": "In this work we present STEVE - Soccer TEam VEctors, a principled approach for learning real valued vectors for soccer teams where similar teams are close to each other in the resulting vector space. STEVE only relies on freely available information about the matches teams played in the past. These vectors can serve as input to various machine learning tasks. Evaluating on the task of team market value estimation, STEVE outperforms all its competitors. Moreover, we use STEVE for similarity search and to rank soccer teams." }, "1908.02269.pdf": { "id": "1908.02269v4", "title": "Promoting Coordination through Policy Regularization in Multi-Agent Deep Reinforcement Learning", "year": "2019", "authors": [ "Julien Roy", "Paul Barde", "F\u00e9lix G. Harvey", "Derek Nowrouzezahrai", "Christopher Pal" ], "categories": "cs.LG, cs.MA, stat.ML", "pdf_url": "https://arxiv.org/pdf/1908.02269v4", "abstract": "In multi-agent reinforcement learning, discovering successful collective behaviors is challenging as it requires exploring a joint action space that grows exponentially with the number of agents. While the tractability of independent agent-wise exploration is appealing, this approach fails on tasks that require elaborate group strategies. We argue that coordinating the agents' policies can guide their exploration and we investigate techniques to promote such an inductive bias. We propose two policy regularization methods: TeamReg, which is based on inter-agent action predictability and CoachReg that relies on synchronized behavior selection. We evaluate each approach on four challenging continuous control tasks with sparse rewards that require varying levels of coordination as well as on the discrete action Google Research Football environment. Our experiments show improved performance across many cooperative multi-agent problems. Finally, we analyze the effects of our proposed methods on the policies that our agents learn and show that our methods successfully enforce the qualities that we propose as proxies for coordinated behaviors." }, "1909.02385.pdf": { "id": "1909.02385v1", "title": "NimbRo Robots Winning RoboCup 2018 Humanoid AdultSize Soccer Competitions", "year": "2019", "authors": [ "Hafez Farazi", "Grzegorz Ficht", "Philipp Allgeuer", "Dmytro Pavlichenko", "Diego Rodriguez", "Andre Brandenburger" ], "categories": "cs.RO", "pdf_url": "https://arxiv.org/pdf/1909.02385v1", "abstract": "Over the past few years, the Humanoid League rules have changed towards more realistic and challenging game environments, which encourage teams to advance their robot soccer performances. In this paper, we present the software and hardware designs that led our team NimbRo to win the competitions in the AdultSize league -- including the soccer tournament, the drop-in games, and the technical challenges at RoboCup 2018 in Montreal. Altogether, this resulted in NimbRo winning the Best Humanoid Award. In particular, we describe our deep-learning approaches for visual perception and our new fully 3D printed robot NimbRo-OP2X." }, "1909.12938.pdf": { "id": "1909.12938v1", "title": "Time Series Modeling for Dream Team in Fantasy Premier League", "year": "2019", "authors": [ "Akhil Gupta" ], "categories": "cs.CY, cs.LG", "pdf_url": "https://arxiv.org/pdf/1909.12938v1", "abstract": "The performance of football players in English Premier League varies largely from season to season and for different teams. It is evident that a method capable of forecasting and analyzing the future of these players on-field antics shall assist the management to a great extent. In a simulated environment like the Fantasy Premier League, enthusiasts from all over the world participate and manage the players catalogue for the entire season. Due to the dynamic nature of points system, there is no known approach for the formulation of a dream team. This study aims to tackle this problem by using a hybrid of Autoregressive Integrated Moving Average (ARIMA) and Recurrent Neural Networks (RNNs) for time series prediction of player points and subsequent maximization of total points using Linear Programming (LPP). Given the player points for the past three seasons, the predictions have been made for the current season by modeling differently for ARIMA and RNN, and then creating an ensemble of the same. Prior to that, proper data preprocessing techniques were deployed to enhance the efficacy of the prepared model. Constraints on the type of players like goalkeepers, defenders, midfielders and forwards along with the total budget were effectively optimized using LPP approach. The validation of the proposed team was done with the performance in upcoming season, where the players outperform as expected, and helped in strengthening the feasibility of the solution. Likewise, the proposed approach can be extended to English Premier League by official managers on-field." }, "1910.00412.pdf": { "id": "1910.00412v1", "title": "\"Does 4-4-2 exist?\" -- An Analytics Approach to Understand and Classify Football Team Formations in Single Match Situations", "year": "2019", "authors": [ "Eric M\u00fcller-Budack", "Jonas Theiner", "Robert Rein", "Ralph Ewerth" ], "categories": "cs.LG, cs.AI", "pdf_url": "https://arxiv.org/pdf/1910.00412v1", "abstract": "The chances to win a football match can be significantly increased if the right tactic is chosen and the behavior of the opposite team is well anticipated. For this reason, every professional football club employs a team of game analysts. However, at present game performance analysis is done manually and therefore highly time-consuming. Consequently, automated tools to support the analysis process are required. In this context, one of the main tasks is to summarize team formations by patterns such as 4-4-2. In this paper, we introduce an analytics approach that automatically classifies and visualizes the team formation based on the players' position data. We focus on single match situations instead of complete halftimes or matches to provide a more detailed analysis. A detailed analysis of individual match situations depending on ball possession and match segment length is provided. For this purpose, a visual summary is utilized that summarizes the team formation in a match segment. An expert annotation study is conducted that demonstrates 1) the complexity of the task and 2) the usefulness of the visualization of single situations to understand team formations. The suggested classification approach outperforms existing methods for formation classification. In particular, our approach gives insights about the shortcomings of using patterns like 4-4-2 to describe team formations." }, "1910.06081.pdf": { "id": "1910.06081v1", "title": "Understanding and Pushing the Limits of the Elo Rating Algorithm", "year": "2019", "authors": [ "Leszek Szczecinski", "Aymen Djebbi" ], "categories": "math.ST", "pdf_url": "https://arxiv.org/pdf/1910.06081v1", "abstract": "This work is concerned with the rating of players/teams in face-to-face games with three possible outcomes: loss, win, and draw. This is one of the fundamental problems in sport analytics, where the very simple and popular, non-trivial algorithm was proposed by Arpad Elo in late fifties to rate chess players. In this work we explain the mathematical model underlying the Elo algorithm and, in particular, we explain what is the implicit but not yet spelled out, assumption about the model of draws. We further extend the model to provide flexibility and remove the unrealistic implicit assumptions of the Elo algorithm. This yields the new rating algorithm, we call $\u03ba$-Elo, which is equally simple as the Elo algorithm but provides a possibility to adjust to the frequency of draws. The discussion of the importance of the appropriate choice of the parameters is carried out and illustrated using results from English Premier League football seasons." }, "1910.06591.pdf": { "id": "1910.06591v2", "title": "SEED RL: Scalable and Efficient Deep-RL with Accelerated Central Inference", "year": "2019", "authors": [ "Lasse Espeholt", "Rapha\u00ebl Marinier", "Piotr Stanczyk", "Ke Wang", "Marcin Michalski" ], "categories": "cs.LG, stat.ML", "pdf_url": "https://arxiv.org/pdf/1910.06591v2", "abstract": "We present a modern scalable reinforcement learning agent called SEED (Scalable, Efficient Deep-RL). By effectively utilizing modern accelerators, we show that it is not only possible to train on millions of frames per second but also to lower the cost of experiments compared to current methods. We achieve this with a simple architecture that features centralized inference and an optimized communication layer. SEED adopts two state of the art distributed algorithms, IMPALA/V-trace (policy gradients) and R2D2 (Q-learning), and is evaluated on Atari-57, DeepMind Lab and Google Research Football. We improve the state of the art on Football and are able to reach state of the art on Atari-57 three times faster in wall-time. For the scenarios we consider, a 40% to 80% cost reduction for running experiments is achieved. The implementation along with experiments is open-sourced so results can be reproduced and novel ideas tried out." }, "1911.04541.pdf": { "id": "1911.04541v5", "title": "Bayesian models for prediction of the set-difference in volleyball", "year": "2019", "authors": [ "Ioannis Ntzoufras", "Vasilis Palaskas", "Sotiris Drikos" ], "categories": "stat.AP", "pdf_url": "https://arxiv.org/pdf/1911.04541v5", "abstract": "The aim of this paper is to study and develop Bayesian models for the analysis of volleyball match outcomes as recorded by the set-difference. Due to the peculiarity of the outcome variable (set-difference) which takes discrete values from $-3$ to $3$, we cannot consider standard models based on the usual Poisson or binomial assumptions used for other sports such as football/soccer. Hence, the first and foremost challenge was to build models appropriate for the set-differences of each volleyball match. Here we consider two major approaches: a) an ordered multinomial logistic regression model and b) a model based on a truncated version of the Skellam distribution. For the first model, we consider the set-difference as an ordinal response variable within the framework of multinomial logistic regression models. Concerning the second model, we adjust the Skellam distribution in order to account for the volleyball rules. We fit and compare both models with the same covariate structure as in Karlis & Ntzoufras (2003). Both models are fitted, illustrated and compared within Bayesian framework using data from both the regular season and the play-offs of the season 2016/17 of the Greek national men's volleyball league A1." }, "1911.04689.pdf": { "id": "1911.04689v2", "title": "Maximizing performance with an eye on the finances: a chance-constrained model for football transfer market decisions", "year": "2019", "authors": [ "Giovanni Pantuso", "Lars Magnus Hvattum" ], "categories": "math.OC", "pdf_url": "https://arxiv.org/pdf/1911.04689v2", "abstract": "Composing a team of professional players is among the most crucial decisions in association football. Nevertheless, transfer market decisions are often based on myopic objectives and are questionable from a financial point of view. This paper introduces a chance-constrained model to provide analytic support to club managers during transfer windows. The model seeks a top-performing team while adapting to different budgets and financial-risk profiles. In addition, it provides a new rating system that is able to numerically reflect the on-field performance of football players and thus contribute to an objective assessment of football players. The model and rating system are tested on a case study based on real market data. The data from the case study are available online for the benefit of future research." }, "1912.01326.pdf": { "id": "1912.01326v3", "title": "A Context-Aware Loss Function for Action Spotting in Soccer Videos", "year": "2019", "authors": [ "Anthony Cioppa", "Adrien Deli\u00e8ge", "Silvio Giancola", "Bernard Ghanem", "Marc Van Droogenbroeck", "Rikke Gade" ], "categories": "cs.CV, cs.LG, eess.IV", "pdf_url": "https://arxiv.org/pdf/1912.01326v3", "abstract": "In video understanding, action spotting consists in temporally localizing human-induced events annotated with single timestamps. In this paper, we propose a novel loss function that specifically considers the temporal context naturally present around each action, rather than focusing on the single annotated frame to spot. We benchmark our loss on a large dataset of soccer videos, SoccerNet, and achieve an improvement of 12.8% over the baseline. We show the generalization capability of our loss for generic activity proposals and detection on ActivityNet, by spotting the beginning and the end of each activity. Furthermore, we provide an extended ablation study and display challenging cases for action spotting in soccer videos. Finally, we qualitatively illustrate how our loss induces a precise temporal understanding of actions and show how such semantic knowledge can be used for automatic highlights generation." }, "1912.07441.pdf": { "id": "1912.07441v1", "title": "Multi-stream Data Analytics for Enhanced Performance Prediction in Fantasy Football", "year": "2019", "authors": [ "Nicholas Bonello", "Joeran Beel", "Seamus Lawless", "Jeremy Debattista" ], "categories": "cs.LG, stat.ML", "pdf_url": "https://arxiv.org/pdf/1912.07441v1", "abstract": "Fantasy Premier League (FPL) performance predictors tend to base their algorithms purely on historical statistical data. The main problems with this approach is that external factors such as injuries, managerial decisions and other tournament match statistics can never be factored into the final predictions. In this paper, we present a new method for predicting future player performances by automatically incorporating human feedback into our model. Through statistical data analysis such as previous performances, upcoming fixture difficulty ratings, betting market analysis, opinions of the general-public and experts alike via social media and web articles, we can improve our understanding of who is likely to perform well in upcoming matches. When tested on the English Premier League 2018/19 season, the model outperformed regular statistical predictors by over 300 points, an average of 11 points per week, ranking within the top 0.5% of players rank 30,000 out of over 6.5 million players." }, "1912.10891.pdf": { "id": "1912.10891v2", "title": "Soft Q Network", "year": "2019", "authors": [ "Jingbin Liu", "Shuai Liu", "Xinyang Gu" ], "categories": "cs.LG, cs.AI", "pdf_url": "https://arxiv.org/pdf/1912.10891v2", "abstract": "Deep Q Network (DQN) is a very successful algorithm, yet the inherent problem of reinforcement learning, i.e. the exploit-explore balance, remains. In this work, we introduce entropy regularization into DQN and propose SQN. We find that the backup equation of soft Q learning can enjoy the corrective feedback if we view the soft backup as policy improvement in the form of Q, instead of policy evaluation. We show that Soft Q Learning with Corrective Feedback (SQL-CF) underlies the on-plicy nature of SQL and the equivalence of SQL and Soft Policy Gradient (SPG). With these insights, we propose an on-policy version of deep Q learning algorithm, i.e. Q On-Policy (QOP). We experiment with QOP on a self-play environment called Google Research Football (GRF). The QOP algorithm exhibits great stability and efficiency in training GRF agents." }, "2001.07692.pdf": { "id": "2001.07692v1", "title": "Estimating locomotor demands during team play from broadcast-derived tracking data", "year": "2020", "authors": [ "Jacob Mortensen", "Luke Bornn" ], "categories": "stat.AP", "pdf_url": "https://arxiv.org/pdf/2001.07692v1", "abstract": "The introduction of optical tracking data across sports has given rise to the ability to dissect athletic performance at a level unfathomable a decade ago. One specific area that has seen substantial benefit is sports science, as high resolution coordinate data permits sports scientists to have to-the-second estimates of external load metrics, such as acceleration load and high speed running distance, traditionally used to understand the physical toll a game takes on an athlete. Unfortunately, collecting this data requires installation of expensive hardware and paying costly licensing fees to data providers, restricting its availability. Algorithms have been developed that allow a traditional broadcast feed to be converted to x-y coordinate data, making tracking data easier to acquire, but coordinates are available for an athlete only when that player is within the camera frame. Obviously, this leads to inaccuracies in player load estimates, limiting the usefulness of this data for sports scientists. In this research, we develop models that predict offscreen load metrics and demonstrate the viability of broadcast-derived tracking data for understanding external load in soccer." }, "2003.01712.pdf": { "id": "2003.01712v1", "title": "Player Chemistry: Striving for a Perfectly Balanced Soccer Team", "year": "2020", "authors": [ "Lotte Bransen", "Jan Van Haaren" ], "categories": "cs.SI, stat.AP", "pdf_url": "https://arxiv.org/pdf/2003.01712v1", "abstract": "Soccer scouts typically ignore the team balance and team chemistry when evaluating potential signings for their teams. Instead, they focus on the individual qualities of the players in isolation. To overcome this limitation of their recruitment process, this paper takes a first step towards objectively providing insight into the question: How well does a team of soccer players gel? We address that question in both an observational and a predictive setting. In the former setting, we observe the chemistry between players who have actually played together, which is relevant when selecting the best possible line-up for a match. In the latter setting, we predict the chemistry between players who have never played together before, which is particularly relevant to assess the fit of a potential signing with the players who are already on the team. We introduce two chemistry metrics that measure the offensive and defensive chemistry for a pair of players, respectively. The offensive chemistry metric measures the pair's joint performance in terms of scoring goals, whereas the defensive chemistry metric measures their joint performance in preventing their opponents from scoring goals. We compute our metrics for 361 seasons in 106 different competitions and present a number of concrete use cases. For instance, we show that the partnership between Mohamed Salah and Roberto Firmino in Liverpool's 2017/2018 Champions League campaign exhibited the highest mutual chemistry between two players. Furthermore, we show that Mesut \u00d6zil's chemistry has rapidly started declining following Alexis S\u00e1nchez' departure to Manchester United in 2018." }, "2003.09384.pdf": { "id": "2003.09384v2", "title": "Investigating the efficiency of the Asian handicap football betting market with ratings and Bayesian networks", "year": "2020", "authors": [ "Anthony Constantinou" ], "categories": "stat.AP, cs.AI", "pdf_url": "https://arxiv.org/pdf/2003.09384v2", "abstract": "Despite the massive popularity of the Asian Handicap (AH) football (soccer) betting market, its efficiency has not been adequately studied by the relevant literature. This paper combines rating systems with Bayesian networks and presents the first published model specifically developed for prediction and assessment of the efficiency of the AH betting market. The results are based on 13 English Premier League seasons and are compared to the traditional market, where the bets are for win, lose or draw. Different betting situations have been examined including a) both average and maximum (best available) market odds, b) all possible betting decision thresholds between predicted and published odds, c) optimisations for both return-on-investment and profit, and d) simple stake adjustments to investigate how the variance of returns changes when targeting equivalent profit in both traditional and AH markets. While the AH market is found to share the inefficiencies of the traditional market, the findings reveal both interesting differences as well as similarities between the two." }, "2003.10294.pdf": { "id": "2003.10294v1", "title": "Optimising Game Tactics for Football", "year": "2020", "authors": [ "Ryan Beal", "Georgios Chalkiadakis", "Timothy J. Norman", "Sarvapali D. Ramchurn" ], "categories": "cs.AI, cs.GT, cs.MA", "pdf_url": "https://arxiv.org/pdf/2003.10294v1", "abstract": "In this paper we present a novel approach to optimise tactical and strategic decision making in football (soccer). We model the game of football as a multi-stage game which is made up from a Bayesian game to model the pre-match decisions and a stochastic game to model the in-match state transitions and decisions. Using this formulation, we propose a method to predict the probability of game outcomes and the payoffs of team actions. Building upon this, we develop algorithms to optimise team formation and in-game tactics with different objectives. Empirical evaluation of our approach on real-world datasets from 760 matches shows that by using optimised tactics from our Bayesian and stochastic games, we can increase a team chances of winning by up to 16.1\\% and 3.4\\% respectively." }, "2003.11102.pdf": { "id": "2003.11102v1", "title": "Learning to Play Soccer by Reinforcement and Applying Sim-to-Real to Compete in the Real World", "year": "2020", "authors": [ "Hansenclever F. Bassani", "Renie A. Delgado", "Jose Nilton de O. Lima Junior", "Heitor R. Medeiros", "Pedro H. M. Braga", "Alain Tapp" ], "categories": "cs.LG, cs.AI, cs.RO, stat.ML", "pdf_url": "https://arxiv.org/pdf/2003.11102v1", "abstract": "This work presents an application of Reinforcement Learning (RL) for the complete control of real soccer robots of the IEEE Very Small Size Soccer (VSSS), a traditional league in the Latin American Robotics Competition (LARC). In the VSSS league, two teams of three small robots play against each other. We propose a simulated environment in which continuous or discrete control policies can be trained, and a Sim-to-Real method to allow using the obtained policies to control a robot in the real world. The results show that the learned policies display a broad repertoire of behaviors that are difficult to specify by hand. This approach, called VSSS-RL, was able to beat the human-designed policy for the striker of the team ranked 3rd place in the 2018 LARC, in 1-vs-1 matches." }, "2004.04147.pdf": { "id": "2004.04147v2", "title": "Slicing and dicing soccer: automatic detection of complex events from spatio-temporal data", "year": "2020", "authors": [ "Lia Morra", "Francesco Manigrasso", "Giuseppe Canto", "Claudio Gianfrate", "Enrico Guarino", "Fabrizio Lamberti" ], "categories": "cs.CV, cs.LO", "pdf_url": "https://arxiv.org/pdf/2004.04147v2", "abstract": "The automatic detection of events in sport videos has im-portant applications for data analytics, as well as for broadcasting andmedia companies. This paper presents a comprehensive approach for de-tecting a wide range of complex events in soccer videos starting frompositional data. The event detector is designed as a two-tier system thatdetectsatomicandcomplex events. Atomic events are detected basedon temporal and logical combinations of the detected objects, their rel-ative distances, as well as spatio-temporal features such as velocity andacceleration. Complex events are defined as temporal and logical com-binations of atomic and complex events, and are expressed by meansof a declarative Interval Temporal Logic (ITL). The effectiveness of theproposed approach is demonstrated over 16 different events, includingcomplex situations such as tackles and filtering passes. By formalizingevents based on principled ITL, it is possible to easily perform reason-ing tasks, such as understanding which passes or crosses result in a goalbeing scored. To counterbalance the lack of suitable, annotated publicdatasets, we built on an open source soccer simulation engine to re-lease the synthetic SoccER (Soccer Event Recognition) dataset, whichincludes complete positional data and annotations for more than 1.6 mil-lion atomic events and 9,000 complex events. The dataset and code areavailable at https://gitlab.com/grains2/slicing-and-dicing-soccer" }, "2004.10299.pdf": { "id": "2004.10299v1", "title": "Group Activity Detection from Trajectory and Video Data in Soccer", "year": "2020", "authors": [ "Ryan Sanford", "Siavash Gorji", "Luiz G. Hafemann", "Bahareh Pourbabaee", "Mehrsan Javan" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2004.10299v1", "abstract": "Group activity detection in soccer can be done by using either video data or player and ball trajectory data. In current soccer activity datasets, activities are labelled as atomic events without a duration. Given that the state-of-the-art activity detection methods are not well-defined for atomic actions, these methods cannot be used. In this work, we evaluated the effectiveness of activity recognition models for detecting such events, by using an intuitive non-maximum suppression process and evaluation metrics. We also considered the problem of explicitly modeling interactions between players and ball. For this, we propose self-attention models to learn and extract relevant information from a group of soccer players for activity detection from both trajectory and video data. We conducted an extensive study on the use of visual features and trajectory data for group activity detection in sports using a large scale soccer dataset provided by Sportlogiq. Our results show that most events can be detected using either vision or trajectory-based approaches with a temporal resolution of less than 0.5 seconds, and that each approach has unique challenges." }, "2006.11909.pdf": { "id": "2006.11909v2", "title": "Two-Sample Testing on Ranked Preference Data and the Role of Modeling Assumptions", "year": "2020", "authors": [ "Charvi Rastogi", "Sivaraman Balakrishnan", "Nihar B. Shah", "Aarti Singh" ], "categories": "stat.ML, cs.IT, cs.LG", "pdf_url": "https://arxiv.org/pdf/2006.11909v2", "abstract": "A number of applications require two-sample testing on ranked preference data. For instance, in crowdsourcing, there is a long-standing question of whether pairwise comparison data provided by people is distributed similar to ratings-converted-to-comparisons. Other examples include sports data analysis and peer grading. In this paper, we design two-sample tests for pairwise comparison data and ranking data. For our two-sample test for pairwise comparison data, we establish an upper bound on the sample complexity required to correctly distinguish between the distributions of the two sets of samples. Our test requires essentially no assumptions on the distributions. We then prove complementary lower bounds showing that our results are tight (in the minimax sense) up to constant factors. We investigate the role of modeling assumptions by proving lower bounds for a range of pairwise comparison models (WST, MST,SST, parameter-based such as BTL and Thurstone). We also provide testing algorithms and associated sample complexity bounds for the problem of two-sample testing with partial (or total) ranking data.Furthermore, we empirically evaluate our results via extensive simulations as well as two real-world datasets consisting of pairwise comparisons. By applying our two-sample test on real-world pairwise comparison data, we conclude that ratings and rankings provided by people are indeed distributed differently. On the other hand, our test recognizes no significant difference in the relative performance of European football teams across two seasons. Finally, we apply our two-sample test on a real-world partial and total ranking dataset and find a statistically significant difference in Sushi preferences across demographic divisions based on gender, age and region of residence." }, "2006.13534.pdf": { "id": "2006.13534v1", "title": "Namira Soccer 2D Simulation Team Description Paper 2020", "year": "2020", "authors": [ "Ehsan Asali", "Farzin Negahbani", "Shahriyar Bamaei", "Zahra Abbasi" ], "categories": "cs.RO, cs.AI, cs.MA", "pdf_url": "https://arxiv.org/pdf/2006.13534v1", "abstract": "In this article, we will discuss methods and ideas which are implemented on Namira 2D Soccer Simulation team in the recent year. Numerous scientific and programming activities were done in the process of code development, but we will mention the most outstanding ones in details. A Kalman filtering method for localization and two helpful software packages will be discussed here. Namira uses agent2d-3.1.1 as base code and librcsc-4.1.0 as library with some deliberate changes." }, "2007.03155.pdf": { "id": "2007.03155v2", "title": "Decentralized policy learning with partial observation and mechanical constraints for multiperson modeling", "year": "2020", "authors": [ "Keisuke Fujii", "Naoya Takeishi", "Yoshinobu Kawahara", "Kazuya Takeda" ], "categories": "cs.LG, cs.MA, stat.ML", "pdf_url": "https://arxiv.org/pdf/2007.03155v2", "abstract": "Extracting the rules of real-world multi-agent behaviors is a current challenge in various scientific and engineering fields. Biological agents independently have limited observation and mechanical constraints; however, most of the conventional data-driven models ignore such assumptions, resulting in lack of biological plausibility and model interpretability for behavioral analyses. Here we propose sequential generative models with partial observation and mechanical constraints in a decentralized manner, which can model agents' cognition and body dynamics, and predict biologically plausible behaviors. We formulate this as a decentralized multi-agent imitation-learning problem, leveraging binary partial observation and decentralized policy models based on hierarchical variational recurrent neural networks with physical and biomechanical penalties. Using real-world basketball and soccer datasets, we show the effectiveness of our method in terms of the constraint violations, long-term trajectory prediction, and partial observation. Our approach can be used as a multi-agent simulator to generate realistic trajectories using real-world data." }, "2007.06475.pdf": { "id": "2007.06475v1", "title": "Automatic Pass Annotation from Soccer VideoStreams Based on Object Detection and LSTM", "year": "2020", "authors": [ "Danilo Sorano", "Fabio Carrara", "Paolo Cintia", "Fabrizio Falchi", "Luca Pappalardo" ], "categories": "cs.CV, cs.LG", "pdf_url": "https://arxiv.org/pdf/2007.06475v1", "abstract": "Soccer analytics is attracting increasing interest in academia and industry, thanks to the availability of data that describe all the spatio-temporal events that occur in each match. These events (e.g., passes, shots, fouls) are collected by human operators manually, constituting a considerable cost for data providers in terms of time and economic resources. In this paper, we describe PassNet, a method to recognize the most frequent events in soccer, i.e., passes, from video streams. Our model combines a set of artificial neural networks that perform feature extraction from video streams, object detection to identify the positions of the ball and the players, and classification of frame sequences as passes or not passes. We test PassNet on different scenarios, depending on the similarity of conditions to the match used for training. Our results show good classification results and significant improvement in the accuracy of pass detection with respect to baseline classifiers, even when the match's video conditions of the test and training sets are considerably different. PassNet is the first step towards an automated event annotation system that may break the time and the costs for event annotation, enabling data collections for minor and non-professional divisions, youth leagues and, in general, competitions whose matches are not currently annotated by data providers." }, "2008.00682.pdf": { "id": "2008.00682v2", "title": "Discovering indicators of dark horse of soccer games by deep learning from sequential trading data", "year": "2020", "authors": [ "Liyao Lu", "Qiang Lyu" ], "categories": "cs.LG, cs.AI", "pdf_url": "https://arxiv.org/pdf/2008.00682v2", "abstract": "It is not surprise for machine learning models to provide decent prediction accuracy of soccer games outcomes based on various objective metrics. However, the performance is not that decent in terms of predicting difficult and valuable matches. A deep learning model is designed and trained on a real sequential trading data from the real prediction market, with the assumption that such trading data contain critical latent information to determine the game outcomes. A new loss function is proposed which biases the selection toward matches with high investment return to train our model. Full investigation of 4669 top soccer league matches showed that our model traded off prediction accuracy for high value return due to a certain ability to detect dark horses. A further try is conducted to depict some indicators discovered by our model for describing key features of big dark horses and regular hot horses." }, "2008.03456.pdf": { "id": "2008.03456v1", "title": "Cyrus 2D Simulation Team Description Paper2018", "year": "2020", "authors": [ "Nader Zare", "Mohsen Sadeghipour", "Ashkan Keshavarzi", "Mahtab Sarvmaili", "Amin Nikanjam", "Reza Aghayari" ], "categories": "cs.RO", "pdf_url": "https://arxiv.org/pdf/2008.03456v1", "abstract": "Cyrus 2D Soccer Simulation was established 2012 with the aim of research and develop in multi agents systems. This year we have joined with Ziziphus for collaboration and speed up our researches. This paper express a brief description of a method for predicting player's behavior in a multi agent system using neural network with a dataset in three level (low, mid, high). The dataset was obtained from log files of past years RoboCup's matches. Behavior Prediction is used in block, mark and defensive decisions. The base code that Cyrus used is agent 3.11." }, "2008.05891.pdf": { "id": "2008.05891v2", "title": "Dodge and survive: modeling the predatory nature of dodgeball", "year": "2020", "authors": [ "Perrin E. Ruth", "Juan G. Restrepo" ], "categories": "physics.soc-ph, nlin.AO", "pdf_url": "https://arxiv.org/pdf/2008.05891v2", "abstract": "The analysis of games and sports as complex systems can give insights into the dynamics of human competition, and has been proven useful in soccer, basketball, and other professional sports. In this paper we present a model for dodgeball, a popular sport in US schools, and analyze it using an ordinary differential equation (ODE) compartmental model and stochastic agent-based game simulations. The ODE model reveals a rich landscape with different game dynamics occurring depending on the strategies used by the teams, which can in some cases be mapped to scenarios in competitive species models. Stochastic agent-based game simulations confirm and complement the predictions of the deterministic ODE models. In some scenarios, game victory can be interpreted as a noise-driven escape from the basin of attraction of a stable fixed point, resulting in extremely long games when the number of players is large. Using the ODE and agent-based models, we construct a strategy to increase the probability of winning." }, "2008.11229.pdf": { "id": "2008.11229v1", "title": "Uncovering Soccer Teams Passing Strategies Using Implication Rules", "year": "2020", "authors": [ "Olumide Leshi" ], "categories": "cs.AI, cs.DB", "pdf_url": "https://arxiv.org/pdf/2008.11229v1", "abstract": "Formal Concept Analysis FCA has seen application in different knowledge areas, including Social Network Analysis SNA. In turn, research has also shown the applicability of SNA in assessing team sports. In this project, to uncover frequent passing sequences of a soccer team, an FCA based approach is introduced. The approach relies on a minimum cover of implications, the Duquenne Guigues DG basis and the notion that a soccer teams passes describe a social network." }, "2008.12624.pdf": { "id": "2008.12624v1", "title": "A Framework for Studying Reinforcement Learning and Sim-to-Real in Robot Soccer", "year": "2020", "authors": [ "Hansenclever F. Bassani", "Renie A. Delgado", "Jos\u00e9 Nilton de O. Lima Junior", "Heitor R. Medeiros", "Pedro H. M. Braga", "Mateus G. Machado" ], "categories": "cs.RO, cs.AI, cs.LG", "pdf_url": "https://arxiv.org/pdf/2008.12624v1", "abstract": "This article introduces an open framework, called VSSS-RL, for studying Reinforcement Learning (RL) and sim-to-real in robot soccer, focusing on the IEEE Very Small Size Soccer (VSSS) league. We propose a simulated environment in which continuous or discrete control policies can be trained to control the complete behavior of soccer agents and a sim-to-real method based on domain adaptation to adapt the obtained policies to real robots. Our results show that the trained policies learned a broad repertoire of behaviors that are difficult to implement with handcrafted control policies. With VSSS-RL, we were able to beat human-designed policies in the 2019 Latin American Robotics Competition (LARC), achieving 4th place out of 21 teams, being the first to apply Reinforcement Learning (RL) successfully in this competition. Both environment and hardware specifications are available open-source to allow reproducibility of our results and further studies." }, "2008.13005.pdf": { "id": "2008.13005v1", "title": "Wisdom of the crowds forecasting the 2018 FIFA Men's World Cup", "year": "2020", "authors": [ "Marco Henrique de Almeida In\u00e1cio", "Rafael Izbicki", "Danilo Louren\u00e7o Lopes", "Luis Ernesto Salasar", "Jo\u00e3o Poloniato", "Marcio Alves Diniz" ], "categories": "stat.AP", "pdf_url": "https://arxiv.org/pdf/2008.13005v1", "abstract": "The FIFA Men's World Cup Tournament (WCT) is the most important football (soccer) competition, attracting worldwide attention. A popular practice among football fans in Brazil is to organize contests in which each participant informs guesses on the final score of each match. The participants are then ranked according to some scoring rule. Inspired by these contests, we created a website to hold an online contest, in which participants were asked for their probabilities on the outcomes of upcoming matches of the WCT. After each round of the tournament, the ranking of all users based on a proper scoring rule were published. This paper studies the performance of some methods intended to extract the wisdom of the crowds, which are aggregated forecasts that uses some or all of the forecasts available. The later methods are compared to simpler forecasting strategies as well as to statistical prediction models. Our findings corroborate the hypothesis that, at least for sporting events, the wisdom of the crowds offers a competitive forecasting strategy. Specifically, some of these strategies were able to achieve high scores in our contest." }, "2009.11676.pdf": { "id": "2009.11676v2", "title": "Eye Movement Feature Classification for Soccer Goalkeeper Expertise Identification in Virtual Reality", "year": "2020", "authors": [ "Benedikt Hosp", "Florian Schultz", "Oliver H\u00f6ner", "Enkelejda Kasneci" ], "categories": "cs.HC, cs.AI, cs.CV, cs.LG", "pdf_url": "https://arxiv.org/pdf/2009.11676v2", "abstract": "The latest research in expertise assessment of soccer players has affirmed the importance of perceptual skills (especially for decision making) by focusing either on high experimental control or on a realistic presentation. To assess the perceptual skills of athletes in an optimized manner, we captured omnidirectional in-field scenes and showed these to 12 expert, 10 intermediate and 13 novice soccer goalkeepers on virtual reality glasses. All scenes were shown from the same natural goalkeeper perspective and ended after the return pass to the goalkeeper. Based on their gaze behavior we classified their expertise with common machine learning techniques. This pilot study shows promising results for objective classification of goalkeepers expertise based on their gaze behaviour and provided valuable insight to inform the design of training systems to enhance perceptual skills of athletes." }, "2010.09467.pdf": { "id": "2010.09467v1", "title": "ARENA: A Data-driven Radio Access Networks Analysis of Football Events", "year": "2020", "authors": [ "Lanfranco Zanzi", "Vincenzo Sciancalepore", "Andres Garcia-Saavedra", "Xavier Costa-Perez", "Georgios Agapiou", "Hans D. Schotten" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2010.09467v1", "abstract": "Mass events represent one of the most challenging scenarios for mobile networks because, although their date and time are usually known in advance, the actual demand for resources is difficult to predict due to its dependency on many different factors. Based on data provided by a major European carrier during mass events in a football stadium comprising up to 30.000 people, 16 base station sectors and $1$Km$^2$ area, we performed a data-driven analysis of the radio access network infrastructure dynamics during such events. Given the insights obtained from the analysis, we developed ARENA, a model-free deep learning Radio Access Network (RAN) capacity forecasting solution that, taking as input past network monitoring data and events context information, provides guidance to mobile operators on the expected RAN capacity needed during a future event. Our results, validated against real events contained in the dataset, illustrate the effectiveness of our proposed solution." }, "2010.10202.pdf": { "id": "2010.10202v1", "title": "SoccerMap: A Deep Learning Architecture for Visually-Interpretable Analysis in Soccer", "year": "2020", "authors": [ "Javier Fern\u00e1ndez", "Luke Bornn" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2010.10202v1", "abstract": "We present a fully convolutional neural network architecture that is capable of estimating full probability surfaces of potential passes in soccer, derived from high-frequency spatiotemporal data. The network receives layers of low-level inputs and learns a feature hierarchy that produces predictions at different sampling levels, capturing both coarse and fine spatial details. By merging these predictions, we can produce visually-rich probability surfaces for any game situation that allows coaches to develop a fine-grained analysis of players' positioning and decision-making, an as-yet little-explored area in sports. We show the network can perform remarkably well in the estimation of pass success probability, and present how it can be adapted easily to approach two other challenging problems: the estimation of pass-selection likelihood and the prediction of the expected value of a pass. Our approach provides a novel solution for learning a full prediction surface when there is only a single-pixel correspondence between ground-truth outcomes and the predicted probability map. The flexibility of this architecture allows its adaptation to a great variety of practical problems in soccer. We also present a set of practical applications, including the evaluation of passing risk at a player level, the identification of the best potential passing options, and the differentiation of passing tendencies between teams." }, "2011.04258.pdf": { "id": "2011.04258v1", "title": "Improved Soccer Action Spotting using both Audio and Video Streams", "year": "2020", "authors": [ "Bastien Vanderplaetse", "St\u00e9phane Dupont" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2011.04258v1", "abstract": "In this paper, we propose a study on multi-modal (audio and video) action spotting and classification in soccer videos. Action spotting and classification are the tasks that consist in finding the temporal anchors of events in a video and determine which event they are. This is an important application of general activity understanding. Here, we propose an experimental study on combining audio and video information at different stages of deep neural network architectures. We used the SoccerNet benchmark dataset, which contains annotated events for 500 soccer game videos from the Big Five European leagues. Through this work, we evaluated several ways to integrate audio stream into video-only-based architectures. We observed an average absolute improvement of the mean Average Precision (mAP) metric of $7.43\\%$ for the action classification task and of $4.19\\%$ for the action spotting task." }, "2011.06014.pdf": { "id": "2011.06014v1", "title": "Football tracking networks: Beyond event-based connectivity", "year": "2020", "authors": [ "J. M. Buldu", "D. Garrido", "D. R. Antequera", "J. Busquets", "E. Estrada", "R. Resta" ], "categories": "cs.SI, physics.soc-ph", "pdf_url": "https://arxiv.org/pdf/2011.06014v1", "abstract": "We propose using Network Science as a complementary tool to analyze player and team behavior during a football match. Specifically, we introduce four kinds of networks based on different ways of interaction between players. Our approach's main novelty is to use tracking datasets to create football tracking networks, instead of constructing and analyzing the traditional networks based on events. In this way, we are able to capture player interactions that go beyond passes and introduce the concepts of (a) Ball Flow Networks, (b) Marking Networks, (c) Signed Proximity Networks and (d) Functional Coordination Networks. After defining the methodology for creating each kind of network, we show some examples using tracking datasets from four different matches of LaLiga Santander. Finally, we discuss some of the applications, limitations, and further improvements of football tracking networks." }, "2011.09192.pdf": { "id": "2011.09192v1", "title": "Game Plan: What AI can do for Football, and What Football can do for AI", "year": "2020", "authors": [ "Karl Tuyls", "Shayegan Omidshafiei", "Paul Muller", "Zhe Wang", "Jerome Connor", "Daniel Hennes" ], "categories": "cs.AI, cs.GT, cs.LG, cs.MA, stat.ML", "pdf_url": "https://arxiv.org/pdf/2011.09192v1", "abstract": "The rapid progress in artificial intelligence (AI) and machine learning has opened unprecedented analytics possibilities in various team and individual sports, including baseball, basketball, and tennis. More recently, AI techniques have been applied to football, due to a huge increase in data collection by professional teams, increased computational power, and advances in machine learning, with the goal of better addressing new scientific challenges involved in the analysis of both individual players' and coordinated teams' behaviors. The research challenges associated with predictive and prescriptive football analytics require new developments and progress at the intersection of statistical learning, game theory, and computer vision. In this paper, we provide an overarching perspective highlighting how the combination of these fields, in particular, forms a unique microcosm for AI research, while offering mutual benefits for professional teams, spectators, and broadcasters in the years to come. We illustrate that this duality makes football analytics a game changer of tremendous value, in terms of not only changing the game of football itself, but also in terms of what this domain can mean for the field of AI. We review the state-of-the-art and exemplify the types of analysis enabled by combining the aforementioned fields, including illustrative examples of counterfactual analysis using predictive models, and the combination of game-theoretic analysis of penalty kicks with statistical learning of player attributes. We conclude by highlighting envisioned downstream impacts, including possibilities for extensions to other sports (real and virtual)." }, "2011.09426.pdf": { "id": "2011.09426v1", "title": "A framework for the fine-grained evaluation of the instantaneous expected value of soccer possessions", "year": "2020", "authors": [ "Javier Fernandez", "Luke Bornn", "Daniel Cervone" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2011.09426v1", "abstract": "The expected possession value (EPV) of a soccer possession represents the likelihood of a team scoring or receiving the next goal at any time instance. By decomposing the EPV into a series of subcomponents that are estimated separately, we develop a comprehensive analysis framework providing soccer practitioners with the ability to evaluate the impact of both observed and potential actions. We show we can obtain calibrated models for all the components of EPV, including a set of yet-unexplored problems in soccer. We produce visually-interpretable probability surfaces for potential passes from a series of deep neural network architectures that learn from low-level spatiotemporal data. Additionally, we present a series of novel practical applications providing coaches with an enriched interpretation of specific game situations." }, "2011.10336.pdf": { "id": "2011.10336v1", "title": "Self-Supervised Small Soccer Player Detection and Tracking", "year": "2020", "authors": [ "Samuel Hurault", "Coloma Ballester", "Gloria Haro" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2011.10336v1", "abstract": "In a soccer game, the information provided by detecting and tracking brings crucial clues to further analyze and understand some tactical aspects of the game, including individual and team actions. State-of-the-art tracking algorithms achieve impressive results in scenarios on which they have been trained for, but they fail in challenging ones such as soccer games. This is frequently due to the player small relative size and the similar appearance among players of the same team. Although a straightforward solution would be to retrain these models by using a more specific dataset, the lack of such publicly available annotated datasets entails searching for other effective solutions. In this work, we propose a self-supervised pipeline which is able to detect and track low-resolution soccer players under different recording conditions without any need of ground-truth data. Extensive quantitative and qualitative experimental results are presented evaluating its performance. We also present a comparison to several state-of-the-art methods showing that both the proposed detector and the proposed tracker achieve top-tier results, in particular in the presence of small players." }, "2011.11785.pdf": { "id": "2011.11785v1", "title": "An analysis of Reinforcement Learning applied to Coach task in IEEE Very Small Size Soccer", "year": "2020", "authors": [ "Carlos H. C. Pena", "Mateus G. Machado", "Mariana S. Barros", "Jos\u00e9 D. P. Silva", "Lucas D. Maciel", "Tsang Ing Ren" ], "categories": "cs.RO, cs.AI, cs.LG", "pdf_url": "https://arxiv.org/pdf/2011.11785v1", "abstract": "The IEEE Very Small Size Soccer (VSSS) is a robot soccer competition in which two teams of three small robots play against each other. Traditionally, a deterministic coach agent will choose the most suitable strategy and formation for each adversary's strategy. Therefore, the role of a coach is of great importance to the game. In this sense, this paper proposes an end-to-end approach for the coaching task based on Reinforcement Learning (RL). The proposed system processes the information during the simulated matches to learn an optimal policy that chooses the current formation, depending on the opponent and game conditions. We trained two RL policies against three different teams (balanced, offensive, and heavily offensive) in a simulated environment. Our results were assessed against one of the top teams of the VSSS league, showing promising results after achieving a win/loss ratio of approximately 2.0." }, "2011.13367.pdf": { "id": "2011.13367v3", "title": "SoccerNet-v2: A Dataset and Benchmarks for Holistic Understanding of Broadcast Soccer Videos", "year": "2020", "authors": [ "Adrien Deli\u00e8ge", "Anthony Cioppa", "Silvio Giancola", "Meisam J. Seikavandi", "Jacob V. Dueholm", "Kamal Nasrollahi" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2011.13367v3", "abstract": "Understanding broadcast videos is a challenging task in computer vision, as it requires generic reasoning capabilities to appreciate the content offered by the video editing. In this work, we propose SoccerNet-v2, a novel large-scale corpus of manual annotations for the SoccerNet video dataset, along with open challenges to encourage more research in soccer understanding and broadcast production. Specifically, we release around 300k annotations within SoccerNet's 500 untrimmed broadcast soccer videos. We extend current tasks in the realm of soccer to include action spotting, camera shot segmentation with boundary detection, and we define a novel replay grounding task. For each task, we provide and discuss benchmark results, reproducible with our open-source adapted implementations of the most relevant works in the field. SoccerNet-v2 is presented to the broader research community to help push computer vision closer to automatic solutions for more general video understanding and production purposes." }, "2012.04380.pdf": { "id": "2012.04380v1", "title": "Combining Machine Learning and Human Experts to Predict Match Outcomes in Football: A Baseline Model", "year": "2020", "authors": [ "Ryan Beal", "Stuart E. Middleton", "Timothy J. Norman", "Sarvapali D. Ramchurn" ], "categories": "cs.CL, cs.AI", "pdf_url": "https://arxiv.org/pdf/2012.04380v1", "abstract": "In this paper, we present a new application-focused benchmark dataset and results from a set of baseline Natural Language Processing and Machine Learning models for prediction of match outcomes for games of football (soccer). By doing so we give a baseline for the prediction accuracy that can be achieved exploiting both statistical match data and contextual articles from human sports journalists. Our dataset is focuses on a representative time-period over 6 seasons of the English Premier League, and includes newspaper match previews from The Guardian. The models presented in this paper achieve an accuracy of 63.18% showing a 6.9% boost on the traditional statistical methods." }, "2012.09849.pdf": { "id": "2012.09849v1", "title": "High-Throughput Synchronous Deep RL", "year": "2020", "authors": [ "Iou-Jen Liu", "Raymond A. Yeh", "Alexander G. Schwing" ], "categories": "cs.LG, cs.AI", "pdf_url": "https://arxiv.org/pdf/2012.09849v1", "abstract": "Deep reinforcement learning (RL) is computationally demanding and requires processing of many data points. Synchronous methods enjoy training stability while having lower data throughput. In contrast, asynchronous methods achieve high throughput but suffer from stability issues and lower sample efficiency due to `stale policies.' To combine the advantages of both methods we propose High-Throughput Synchronous Deep Reinforcement Learning (HTS-RL). In HTS-RL, we perform learning and rollouts concurrently, devise a system design which avoids `stale policies' and ensure that actors interact with environment replicas in an asynchronous manner while maintaining full determinism. We evaluate our approach on Atari games and the Google Research Football environment. Compared to synchronous baselines, HTS-RL is 2-6$\\times$ faster. Compared to state-of-the-art asynchronous methods, HTS-RL has competitive throughput and consistently achieves higher average episode rewards." }, "2012.10531.pdf": { "id": "2012.10531v1", "title": "A Graph Attention Based Approach for Trajectory Prediction in Multi-agent Sports Games", "year": "2020", "authors": [ "Ding Ding", "H. Howie Huang" ], "categories": "cs.LG, cs.AI", "pdf_url": "https://arxiv.org/pdf/2012.10531v1", "abstract": "This work investigates the problem of multi-agents trajectory prediction. Prior approaches lack of capability of capturing fine-grained dependencies among coordinated agents. In this paper, we propose a spatial-temporal trajectory prediction approach that is able to learn the strategy of a team with multiple coordinated agents. In particular, we use graph-based attention model to learn the dependency of the agents. In addition, instead of utilizing the recurrent networks (e.g., VRNN, LSTM), our method uses a Temporal Convolutional Network (TCN) as the sequential model to support long effective history and provide important features such as parallelism and stable gradients. We demonstrate the validation and effectiveness of our approach on two different sports game datasets: basketball and soccer datasets. The result shows that compared to related approaches, our model that infers the dependency of players yields substantially improved performance. Code is available at https://github.com/iHeartGraph/predict" }, "2012.14949.pdf": { "id": "2012.14949v2", "title": "Estimating the change in soccer's home advantage during the Covid-19 pandemic using bivariate Poisson regression", "year": "2020", "authors": [ "Luke S. Benz", "Michael J. Lopez" ], "categories": "stat.AP", "pdf_url": "https://arxiv.org/pdf/2012.14949v2", "abstract": "In wake of the Covid-19 pandemic, 2019-2020 soccer seasons across the world were postponed and eventually made up during the summer months of 2020. Researchers from a variety of disciplines jumped at the opportunity to compare the rescheduled games, played in front of empty stadia, to previous games, played in front of fans. To date, most of this post-Covid soccer research has used linear regression models, or versions thereof, to estimate potential changes to the home advantage. But because soccer outcomes are non-linear, we argue that leveraging the Poisson distribution would be more appropriate. We begin by using simulations to show that bivariate Poisson regression reduces absolute bias when estimating the home advantage benefit in a single season of soccer games, relative to linear regression, by almost 85 percent. Next, with data from 17 professional soccer leagues, we extend bivariate Poisson models estimate the change in home advantage due to games being played without fans. In contrast to current research that overwhelmingly suggests a drop in the home advantage, our findings are mixed; in some leagues, evidence points to a decrease, while in others, the home advantage may have risen. Altogether, this suggests a more complex causal mechanism for the impact of fans on sporting events." }, "2101.05388.pdf": { "id": "2101.05388v1", "title": "Evaluating Soccer Player: from Live Camera to Deep Reinforcement Learning", "year": "2021", "authors": [ "Paul Garnier", "Th\u00e9ophane Gregoir" ], "categories": "cs.LG, cs.GT", "pdf_url": "https://arxiv.org/pdf/2101.05388v1", "abstract": "Scientifically evaluating soccer players represents a challenging Machine Learning problem. Unfortunately, most existing answers have very opaque algorithm training procedures; relevant data are scarcely accessible and almost impossible to generate. In this paper, we will introduce a two-part solution: an open-source Player Tracking model and a new approach to evaluate these players based solely on Deep Reinforcement Learning, without human data training nor guidance. Our tracking model was trained in a supervised fashion on datasets we will also release, and our Evaluation Model relies only on simulations of virtual soccer games. Combining those two architectures allows one to evaluate Soccer Players directly from a live camera without large datasets constraints. We term our new approach Expected Discounted Goal (EDG), as it represents the number of goals a team can score or concede from a particular state. This approach leads to more meaningful results than the existing ones that are based on real-world data, and could easily be extended to other sports." }, "2102.04331.pdf": { "id": "2102.04331v1", "title": "Soccer Event Detection Using Deep Learning", "year": "2021", "authors": [ "Ali Karimi", "Ramin Toosi", "Mohammad Ali Akhaee" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2102.04331v1", "abstract": "Event detection is an important step in extracting knowledge from the video. In this paper, we propose a deep learning approach to detect events in a soccer match emphasizing the distinction between images of red and yellow cards and the correct detection of the images of selected events from other images. This method includes the following three modules: i) the variational autoencoder (VAE) module to differentiate between soccer images and others image, ii) the image classification module to classify the images of events, and iii) the fine-grain image classification module to classify the images of red and yellow cards. Additionally, a new dataset was introduced for soccer images classification that is employed to train the networks mentioned in the paper. In the final section, 10 UEFA Champions League matches are used to evaluate the networks' performance and precision in detecting the events. The experiments demonstrate that the proposed method achieves better performance than state-of-the-art methods." }, "2102.07545.pdf": { "id": "2102.07545v2", "title": "Data-driven Analysis for Understanding Team Sports Behaviors", "year": "2021", "authors": [ "Keisuke Fujii" ], "categories": "cs.AI", "pdf_url": "https://arxiv.org/pdf/2102.07545v2", "abstract": "Understanding the principles of real-world biological multi-agent behaviors is a current challenge in various scientific and engineering fields. The rules regarding the real-world biological multi-agent behaviors such as team sports are often largely unknown due to their inherently higher-order interactions, cognition, and body dynamics. Estimation of the rules from data, i.e., data-driven approaches such as machine learning, provides an effective way for the analysis of such behaviors. Although most data-driven models have non-linear structures and high prediction performances, it is sometimes hard to interpret them. This survey focuses on data-driven analysis for quantitative understanding of invasion team sports behaviors such as basketball and football, and introduces two main approaches for understanding such multi-agent behaviors: (1) extracting easily interpretable features or rules from data and (2) generating and controlling behaviors in visually-understandable ways. The first approach involves the visualization of learned representations and the extraction of mathematical structures behind the behaviors. The second approach can be used to test hypotheses by simulating and controlling future and counterfactual behaviors. Lastly, the potential practical applications of extracted rules, features, and generated behaviors are discussed. These approaches can contribute to a better understanding of multi-agent behaviors in the real world." }, "2102.07624.pdf": { "id": "2102.07624v1", "title": "RMS-Net: Regression and Masking for Soccer Event Spotting", "year": "2021", "authors": [ "Matteo Tomei", "Lorenzo Baraldi", "Simone Calderara", "Simone Bronzin", "Rita Cucchiara" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2102.07624v1", "abstract": "The recently proposed action spotting task consists in finding the exact timestamp in which an event occurs. This task fits particularly well for soccer videos, where events correspond to salient actions strictly defined by soccer rules (a goal occurs when the ball crosses the goal line). In this paper, we devise a lightweight and modular network for action spotting, which can simultaneously predict the event label and its temporal offset using the same underlying features. We enrich our model with two training strategies: the first one for data balancing and uniform sampling, the second for masking ambiguous frames and keeping the most discriminative visual cues. When tested on the SoccerNet dataset and using standard features, our full proposal exceeds the current state of the art by 3 Average-mAP points. Additionally, it reaches a gain of more than 10 Average-mAP points on the test set when fine-tuned in combination with a strong 2D backbone." }, "2102.09381.pdf": { "id": "2102.09381v1", "title": "L2E: Learning to Exploit Your Opponent", "year": "2021", "authors": [ "Zhe Wu", "Kai Li", "Enmin Zhao", "Hang Xu", "Meng Zhang", "Haobo Fu" ], "categories": "cs.LG, cs.AI, cs.GT", "pdf_url": "https://arxiv.org/pdf/2102.09381v1", "abstract": "Opponent modeling is essential to exploit sub-optimal opponents in strategic interactions. Most previous works focus on building explicit models to directly predict the opponents' styles or strategies, which require a large amount of data to train the model and lack adaptability to unknown opponents. In this work, we propose a novel Learning to Exploit (L2E) framework for implicit opponent modeling. L2E acquires the ability to exploit opponents by a few interactions with different opponents during training, thus can adapt to new opponents with unknown styles during testing quickly. We propose a novel opponent strategy generation algorithm that produces effective opponents for training automatically. We evaluate L2E on two poker games and one grid soccer game, which are the commonly used benchmarks for opponent modeling. Comprehensive experimental results indicate that L2E quickly adapts to diverse styles of unknown opponents." }, "2103.01955.pdf": { "id": "2103.01955v4", "title": "The Surprising Effectiveness of PPO in Cooperative, Multi-Agent Games", "year": "2021", "authors": [ "Chao Yu", "Akash Velu", "Eugene Vinitsky", "Jiaxuan Gao", "Yu Wang", "Alexandre Bayen" ], "categories": "cs.LG, cs.AI, cs.MA", "pdf_url": "https://arxiv.org/pdf/2103.01955v4", "abstract": "Proximal Policy Optimization (PPO) is a ubiquitous on-policy reinforcement learning algorithm but is significantly less utilized than off-policy learning algorithms in multi-agent settings. This is often due to the belief that PPO is significantly less sample efficient than off-policy methods in multi-agent systems. In this work, we carefully study the performance of PPO in cooperative multi-agent settings. We show that PPO-based multi-agent algorithms achieve surprisingly strong performance in four popular multi-agent testbeds: the particle-world environments, the StarCraft multi-agent challenge, Google Research Football, and the Hanabi challenge, with minimal hyperparameter tuning and without any domain-specific algorithmic modifications or architectures. Importantly, compared to competitive off-policy methods, PPO often achieves competitive or superior results in both final returns and sample efficiency. Finally, through ablation studies, we analyze implementation and hyperparameter factors that are critical to PPO's empirical performance, and give concrete practical suggestions regarding these factors. Our results show that when using these practices, simple PPO-based methods can be a strong baseline in cooperative multi-agent reinforcement learning. Source code is released at \\url{https://github.com/marlbenchmark/on-policy}." }, "2103.02938.pdf": { "id": "2103.02938v1", "title": "FootApp: an AI-Powered System for Football Match Annotation", "year": "2021", "authors": [ "Silvio Barra", "Salvatore M. Carta", "Alessandro Giuliani", "Alessia Pisu", "Alessandro Sebastian Podda", " DanieleRiboni" ], "categories": "cs.HC, eess.SY", "pdf_url": "https://arxiv.org/pdf/2103.02938v1", "abstract": "In the last years, scientific and industrial research has experienced a growing interest in acquiring large annotated data sets to train artificial intelligence algorithms for tackling problems in different domains. In this context, we have observed that even the market for football data has substantially grown. The analysis of football matches relies on the annotation of both individual players' and team actions, as well as the athletic performance of players. Consequently, annotating football events at a fine-grained level is a very expensive and error-prone task. Most existing semi-automatic tools for football match annotation rely on cameras and computer vision. However, those tools fall short in capturing team dynamics, and in extracting data of players who are not visible in the camera frame. To address these issues, in this manuscript we present FootApp, an AI-based system for football match annotation. First, our system relies on an advanced and mixed user interface that exploits both vocal and touch interaction. Second, the motor performance of players is captured and processed by applying machine learning algorithms to data collected from inertial sensors worn by players. Artificial intelligence techniques are then used to check the consistency of generated labels, including those regarding the physical activity of players, to automatically recognize annotation errors. Notably, we implemented a full prototype of the proposed system, performing experiments to show its effectiveness in a real-world adoption scenario." }, "2103.04647.pdf": { "id": "2103.04647v3", "title": "Flexible marked spatio-temporal point processes with applications to event sequences from association football", "year": "2021", "authors": [ "Santhosh Narayanan", "Ioannis Kosmidis", "Petros Dellaportas" ], "categories": "stat.AP, stat.ME", "pdf_url": "https://arxiv.org/pdf/2103.04647v3", "abstract": "We develop a new family of marked point processes by focusing the characteristic properties of marked Hawkes processes exclusively to the space of marks, providing the freedom to specify a different model for the occurrence times. This is possible through the decomposition of the joint distribution of marks and times that allows to separately specify the conditional distribution of marks given the filtration of the process and the current time. We develop a Bayesian framework for the inference and prediction from this family of marked point processes that can naturally accommodate process and point-specific covariate information to drive cross-excitations, offering wide flexibility and applicability in the modelling of real-world processes. The framework is used here for the modelling of in-game event sequences from association football, resulting not only in inferences about previously unquantified characteristics of game dynamics and extraction of event-specific team abilities, but also in predictions for the occurrence of events of interest, such as goals, corners or fouls in a specified interval of time." }, "2103.05174.pdf": { "id": "2103.05174v1", "title": "Learning to Play Soccer From Scratch: Sample-Efficient Emergent Coordination through Curriculum-Learning and Competition", "year": "2021", "authors": [ "Pavan Samtani", "Francisco Leiva", "Javier Ruiz-del-Solar" ], "categories": "cs.LG, cs.MA", "pdf_url": "https://arxiv.org/pdf/2103.05174v1", "abstract": "This work proposes a scheme that allows learning complex multi-agent behaviors in a sample efficient manner, applied to 2v2 soccer. The problem is formulated as a Markov game, and solved using deep reinforcement learning. We propose a basic multi-agent extension of TD3 for learning the policy of each player, in a decentralized manner. To ease learning, the task of 2v2 soccer is divided in three stages: 1v0, 1v1 and 2v2. The process of learning in multi-agent stages (1v1 and 2v2) uses agents trained on a previous stage as fixed opponents. In addition, we propose using experience sharing, a method that shares experience from a fixed opponent, trained in a previous stage, for training the agent currently learning, and a form of frame-skipping, to raise performance significantly. Our results show that high quality soccer play can be obtained with our approach in just under 40M interactions. A summarized video of the resulting game play can be found in https://youtu.be/f25l1j1U9RM." }, "2103.06939.pdf": { "id": "2103.06939v1", "title": "A Reinforcement Learning Based Approach to Play Calling in Football", "year": "2021", "authors": [ "Preston Biro", "Stephen G. Walker" ], "categories": "cs.LG, stat.AP, stat.ME, stat.ML", "pdf_url": "https://arxiv.org/pdf/2103.06939v1", "abstract": "With the vast amount of data collected on football and the growth of computing abilities, many games involving decision choices can be optimized. The underlying rule is the maximization of an expected utility of outcomes and the law of large numbers. The data available allows us to compute with high accuracy the probabilities of outcomes of decisions and the well defined points system in the game allows us to have the necessary terminal utilities. With some well established theory we can then optimize choices at a single play level." }, "2103.09627.pdf": { "id": "2103.09627v3", "title": "Evaluation of soccer team defense based on prediction models of ball recovery and being attacked: A pilot study", "year": "2021", "authors": [ "Kosuke Toda", "Masakiyo Teranishi", "Keisuke Kushiro", "Keisuke Fujii" ], "categories": "cs.AI", "pdf_url": "https://arxiv.org/pdf/2103.09627v3", "abstract": "With the development of measurement technology, data on the movements of actual games in various sports can be obtained and used for planning and evaluating the tactics and strategy. Defense in team sports is generally difficult to be evaluated because of the lack of statistical data. Conventional evaluation methods based on predictions of scores are considered unreliable because they predict rare events throughout the game. Besides, it is difficult to evaluate various plays leading up to a score. In this study, we propose a method to evaluate team defense from a comprehensive perspective related to team performance by predicting ball recovery and being attacked, which occur more frequently than goals, using player actions and positional data of all players and the ball. Using data from 45 soccer matches, we examined the relationship between the proposed index and team performance in actual matches and throughout a season. Results show that the proposed classifiers predicted the true events (mean F1 score $>$ 0.483) better than the existing classifiers which were based on rare events or goals (mean F1 score $<$ 0.201). Also, the proposed index had a moderate correlation with the long-term outcomes of the season ($r =$ 0.397). These results suggest that the proposed index might be a more reliable indicator rather than winning or losing with the inclusion of accidental factors." }, "2104.03252.pdf": { "id": "2104.03252v2", "title": "Leaving Goals on the Pitch: Evaluating Decision Making in Soccer", "year": "2021", "authors": [ "Maaike Van Roy", "Pieter Robberechts", "Wen-Chi Yang", "Luc De Raedt", "Jesse Davis" ], "categories": "cs.AI", "pdf_url": "https://arxiv.org/pdf/2104.03252v2", "abstract": "Analysis of the popular expected goals (xG) metric in soccer has determined that a (slightly) smaller number of high-quality attempts will likely yield more goals than a slew of low-quality ones. This observation has driven a change in shooting behavior. Teams are passing up on shots from outside the penalty box, in the hopes of generating a better shot closer to goal later on. This paper evaluates whether this decrease in long-distance shots is warranted. Therefore, we propose a novel generic framework to reason about decision-making in soccer by combining techniques from machine learning and artificial intelligence (AI). First, we model how a team has behaved offensively over the course of two seasons by learning a Markov Decision Process (MDP) from event stream data. Second, we use reasoning techniques arising from the AI literature on verification to each team's MDP. This allows us to reason about the efficacy of certain potential decisions by posing counterfactual questions to the MDP. Our key conclusion is that teams would score more goals if they shot more often from outside the penalty box in a small number of team-specific locations. The proposed framework can easily be extended and applied to analyze other aspects of the game." }, "2104.05931.pdf": { "id": "2104.05931v1", "title": "Two-stage training algorithm for AI robot soccer", "year": "2021", "authors": [ "Taeyoung Kim", "Luiz Felipe Vecchietti", "Kyujin Choi", "Sanem Sariel", "Dongsoo Har" ], "categories": "cs.AI", "pdf_url": "https://arxiv.org/pdf/2104.05931v1", "abstract": "In multi-agent reinforcement learning, the cooperative learning behavior of agents is very important. In the field of heterogeneous multi-agent reinforcement learning, cooperative behavior among different types of agents in a group is pursued. Learning a joint-action set during centralized training is an attractive way to obtain such cooperative behavior, however, this method brings limited learning performance with heterogeneous agents. To improve the learning performance of heterogeneous agents during centralized training, two-stage heterogeneous centralized training which allows the training of multiple roles of heterogeneous agents is proposed. During training, two training processes are conducted in a series. One of the two stages is to attempt training each agent according to its role, aiming at the maximization of individual role rewards. The other is for training the agents as a whole to make them learn cooperative behaviors while attempting to maximize shared collective rewards, e.g., team rewards. Because these two training processes are conducted in a series in every timestep, agents can learn how to maximize role rewards and team rewards simultaneously. The proposed method is applied to 5 versus 5 AI robot soccer for validation. Simulation results show that the proposed method can train the robots of the robot soccer team effectively, achieving higher role rewards and higher team rewards as compared to other approaches that can be used to solve problems of training cooperative multi-agent." }, "2104.06779.pdf": { "id": "2104.06779v1", "title": "Temporally-Aware Feature Pooling for Action Spotting in Soccer Broadcasts", "year": "2021", "authors": [ "Silvio Giancola", "Bernard Ghanem" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2104.06779v1", "abstract": "Toward the goal of automatic production for sports broadcasts, a paramount task consists in understanding the high-level semantic information of the game in play. For instance, recognizing and localizing the main actions of the game would allow producers to adapt and automatize the broadcast production, focusing on the important details of the game and maximizing the spectator engagement. In this paper, we focus our analysis on action spotting in soccer broadcast, which consists in temporally localizing the main actions in a soccer game. To that end, we propose a novel feature pooling method based on NetVLAD, dubbed NetVLAD++, that embeds temporally-aware knowledge. Different from previous pooling methods that consider the temporal context as a single set to pool from, we split the context before and after an action occurs. We argue that considering the contextual information around the action spot as a single entity leads to a sub-optimal learning for the pooling module. With NetVLAD++, we disentangle the context from the past and future frames and learn specific vocabularies of semantics for each subsets, avoiding to blend and blur such vocabulary in time. Injecting such prior knowledge creates more informative pooling modules and more discriminative pooled features, leading into a better understanding of the actions. We train and evaluate our methodology on the recent large-scale dataset SoccerNet-v2, reaching 53.4% Average-mAP for action spotting, a +12.7% improvement w.r.t the current state-of-the-art." }, "2104.09402.pdf": { "id": "2104.09402v1", "title": "Agent-Centric Representations for Multi-Agent Reinforcement Learning", "year": "2021", "authors": [ "Wenling Shang", "Lasse Espeholt", "Anton Raichuk", "Tim Salimans" ], "categories": "cs.LG, cs.AI", "pdf_url": "https://arxiv.org/pdf/2104.09402v1", "abstract": "Object-centric representations have recently enabled significant progress in tackling relational reasoning tasks. By building a strong object-centric inductive bias into neural architectures, recent efforts have improved generalization and data efficiency of machine learning algorithms for these problems. One problem class involving relational reasoning that still remains under-explored is multi-agent reinforcement learning (MARL). Here we investigate whether object-centric representations are also beneficial in the fully cooperative MARL setting. Specifically, we study two ways of incorporating an agent-centric inductive bias into our RL algorithm: 1. Introducing an agent-centric attention module with explicit connections across agents 2. Adding an agent-centric unsupervised predictive objective (i.e. not using action labels), to be used as an auxiliary loss for MARL, or as the basis of a pre-training step. We evaluate these approaches on the Google Research Football environment as well as DeepMind Lab 2D. Empirically, agent-centric representation learning leads to the emergence of more complex cooperation strategies between agents as well as enhanced sample efficiency and generalization." }, "2104.10125.pdf": { "id": "2104.10125v1", "title": "Bisecting for selecting: using a Laplacian eigenmaps clustering approach to create the new European football Super League", "year": "2021", "authors": [ "A. J. Bond", "C. B. Beggs" ], "categories": "stat.AP, stat.ME, stat.ML, stat.OT", "pdf_url": "https://arxiv.org/pdf/2104.10125v1", "abstract": "We use European football performance data to select teams to form the proposed European football Super League, using only unsupervised techniques. We first used random forest regression to select important variables predicting goal difference, which we used to calculate the Euclidian distances between teams. Creating a Laplacian eigenmap, we bisected the Fielder vector to identify the five major European football leagues' natural clusters. Our results showed how an unsupervised approach could successfully identify four clusters based on five basic performance metrics: shots, shots on target, shots conceded, possession, and pass success. The top two clusters identify those teams who dominate their respective leagues and are the best candidates to create the most competitive elite super league." }, "2104.10272.pdf": { "id": "2104.10272v2", "title": "Stochastic model for football's collective dynamics", "year": "2021", "authors": [ "A. Chacoma", "N. Almeira", "J. I. Perotti", "O. V. Billoni" ], "categories": "physics.soc-ph", "pdf_url": "https://arxiv.org/pdf/2104.10272v2", "abstract": "In this paper, we study collective interaction dynamics emerging in the game of football-soccer. To do so, we surveyed a database containing body-sensors traces measured during three professional football matches, where we observed statistical patterns that we used to propose a stochastic model for the players' motion in the field. The model, which is based on linear interactions, captures in good approximation the spatiotemporal dynamics of a football team. Our theoretical framework, therefore, becomes an effective analytical tool to uncover the underlying cooperative mechanisms behind the complexity of football plays. Moreover, we showed that it can provide handy theoretical support for coaches to evaluate teams' and players' performances in both training sessions and competitive scenarios." }, "2105.00499.pdf": { "id": "2105.00499v1", "title": "Curious Exploration and Return-based Memory Restoration for Deep Reinforcement Learning", "year": "2021", "authors": [ "Saeed Tafazzol", "Erfan Fathi", "Mahdi Rezaei", "Ehsan Asali" ], "categories": "cs.RO, cs.AI, cs.LG, cs.MA", "pdf_url": "https://arxiv.org/pdf/2105.00499v1", "abstract": "Reward engineering and designing an incentive reward function are non-trivial tasks to train agents in complex environments. Furthermore, an inaccurate reward function may lead to a biased behaviour which is far from an efficient and optimised behaviour. In this paper, we focus on training a single agent to score goals with binary success/failure reward function in Half Field Offense domain. As the major advantage of this research, the agent has no presumption about the environment which means it only follows the original formulation of reinforcement learning agents. The main challenge of using such a reward function is the high sparsity of positive reward signals. To address this problem, we use a simple prediction-based exploration strategy (called Curious Exploration) along with a Return-based Memory Restoration (RMR) technique which tends to remember more valuable memories. The proposed method can be utilized to train agents in environments with fairly complex state and action spaces. Our experimental results show that many recent solutions including our baseline method fail to learn and perform in complex soccer domain. However, the proposed method can converge easily to the nearly optimal behaviour. The video presenting the performance of our trained agent is available at http://bit.ly/HFO_Binary_Reward." }, "2105.00839.pdf": { "id": "2105.00839v1", "title": "Elo Ratings for Large Tournaments of Software Agents in Asymmetric Games", "year": "2021", "authors": [ "Ben Wise" ], "categories": "cs.AI, cs.GT", "pdf_url": "https://arxiv.org/pdf/2105.00839v1", "abstract": "The Elo rating system has been used world wide for individual sports and team sports, as exemplified by the European Go Federation (EGF), International Chess Federation (FIDE), International Federation of Association Football (FIFA), and many others. To evaluate the performance of artificial intelligence agents, it is natural to evaluate them on the same Elo scale as humans, such as the rating of 5185 attributed to AlphaGo Zero. There are several fundamental differences between humans and AI that suggest modifications to the system, which in turn require revisiting Elo's fundamental rationale. AI is typically trained on many more games than humans play, and we have little a-priori information on newly created AI agents. Further, AI is being extended into games which are asymmetric between the players, and which could even have large complex boards with different setup in every game, such as commercial paper strategy games. We present a revised rating system, and guidelines for tournaments, to reflect these differences." }, "2105.00931.pdf": { "id": "2105.00931v2", "title": "GridToPix: Training Embodied Agents with Minimal Supervision", "year": "2021", "authors": [ "Unnat Jain", "Iou-Jen Liu", "Svetlana Lazebnik", "Aniruddha Kembhavi", "Luca Weihs", "Alexander Schwing" ], "categories": "cs.CV, cs.AI, cs.LG, cs.MA", "pdf_url": "https://arxiv.org/pdf/2105.00931v2", "abstract": "While deep reinforcement learning (RL) promises freedom from hand-labeled data, great successes, especially for Embodied AI, require significant work to create supervision via carefully shaped rewards. Indeed, without shaped rewards, i.e., with only terminal rewards, present-day Embodied AI results degrade significantly across Embodied AI problems from single-agent Habitat-based PointGoal Navigation (SPL drops from 55 to 0) and two-agent AI2-THOR-based Furniture Moving (success drops from 58% to 1%) to three-agent Google Football-based 3 vs. 1 with Keeper (game score drops from 0.6 to 0.1). As training from shaped rewards doesn't scale to more realistic tasks, the community needs to improve the success of training with terminal rewards. For this we propose GridToPix: 1) train agents with terminal rewards in gridworlds that generically mirror Embodied AI environments, i.e., they are independent of the task; 2) distill the learned policy into agents that reside in complex visual worlds. Despite learning from only terminal rewards with identical models and RL algorithms, GridToPix significantly improves results across tasks: from PointGoal Navigation (SPL improves from 0 to 64) and Furniture Moving (success improves from 1% to 25%) to football gameplay (game score improves from 0.1 to 0.6). GridToPix even helps to improve the results of shaped reward training." }, "2105.01226.pdf": { "id": "2105.01226v1", "title": "Modelling age-related changes in executive functions of soccer players", "year": "2021", "authors": [ "Vincent Chin", "Adam Beavan", "Job Fransen", "Jan Mayer", "Robert Kohn", "Louise M. Ryan" ], "categories": "stat.AP", "pdf_url": "https://arxiv.org/pdf/2105.01226v1", "abstract": "The widespread popularity of soccer across the globe has turned it into a multi-billion dollar industry. As a result, most professional clubs actively engage in talent identification and development programmes. Contemporary research has generally supported the use of executive functions - a class of neuropsychological processes responsible for cognitive behaviours - in predicting a soccer player's future success. However, studies on the developmental evolution of executive functions have yielded differing results in their structural form (such as inverted U-shapes, or otherwise). This article presents the first analysis of changes in the domain-generic and domain-specific executive functions based on longitudinal data measured on elite German soccer players. Results obtained from a latent variable model show that these executive functions experience noticeable growth from late childhood until pre-adolescence, but remain fairly stable in later growth stages. As a consequence, our results suggest that the role of executive functions in facilitating talent identification may have been overly emphasised." }, "2105.04293.pdf": { "id": "2105.04293v2", "title": "An interactive dashboard for searching and comparing soccer performance scores", "year": "2021", "authors": [ "Paolo Cintia", "Giovanni Mauro", "Luca Pappalardo", "Paolo Ferragina" ], "categories": "cs.HC, cs.AI, cs.IR", "pdf_url": "https://arxiv.org/pdf/2105.04293v2", "abstract": "The performance of soccer players is one of most discussed aspects by many actors in the soccer industry: from supporters to journalists, from coaches to talent scouts. Unfortunately, the dashboards available online provide no effective way to compare the evolution of the performance of players or to find players behaving similarly on the field. This paper describes the design of a web dashboard that interacts via APIs with a performance evaluation algorithm and provides graphical tools that allow the user to perform many tasks, such as to search or compare players by age, role or trend of growth in their performance, find similar players based on their pitching behavior, change the algorithm's parameters to obtain customized performance scores. We also describe an example of how a talent scout can interact with the dashboard to find young, promising talents." }, "2105.04875.pdf": { "id": "2105.04875v1", "title": "Video-based Analysis of Soccer Matches", "year": "2021", "authors": [ "Maximilian T. Fischer", "Daniel A. Keim", "Manuel Stein" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2105.04875v1", "abstract": "With the increasingly detailed investigation of game play and tactics in invasive team sports such as soccer, it becomes ever more important to present causes, actions and findings in a meaningful manner. Visualizations, especially when augmenting relevant information directly inside a video recording of a match, can significantly improve and simplify soccer match preparation and tactic planning. However, while many visualization techniques for soccer have been developed in recent years, few have been directly applied to the video-based analysis of soccer matches. This paper provides a comprehensive overview and categorization of the methods developed for the video-based visual analysis of soccer matches. While identifying the advantages and disadvantages of the individual approaches, we identify and discuss open research questions, soon enabling analysts to develop winning strategies more efficiently, do rapid failure analysis or identify weaknesses in opposing teams." }, "2105.09881.pdf": { "id": "2105.09881v2", "title": "Poisson Modeling and Predicting English Premier League Goal Scoring", "year": "2021", "authors": [ "Quang Nguyen" ], "categories": "stat.AP", "pdf_url": "https://arxiv.org/pdf/2105.09881v2", "abstract": "The English Premier League is well-known for being not only one of the most popular professional sports leagues in the world, but also one of the toughest competitions to predict. The first purpose of this research was to verify the consistency between goal scoring in the English Premier League and the Poisson process; specifically, the relationships between the number of goals scored in a match and the Poisson distribution, the time between goals throughout the course of a season and the exponential distribution, and the time location of goals during football games and the continuous uniform distribution. We found that the Poisson process and the three probability distributions accurately describe Premier League goal scoring. In addition, Poisson regression was utilized to predict outcomes for a Premier League season, using different sets of season data and with a large number of simulations being involved. We examined and compared various soccer metrics from our simulation results, including an English club's chances of being the champions, finishing in the top four and bottom three, and relegation points." }, "2105.10700.pdf": { "id": "2105.10700v1", "title": "Soccer Player Tracking in Low Quality Video", "year": "2021", "authors": [ "Eloi Martins", "Jos\u00e9 Henrique Brito" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2105.10700v1", "abstract": "In this paper we propose a system capable of tracking multiple soccer players in different types of video quality. The main goal, in contrast to most state-of-art soccer player tracking systems, is the ability of execute effectively tracking in videos of low-quality. We adapted a state-of-art Multiple Object Tracking to the task. In order to do that adaptation, we created a Detection and a Tracking Dataset for 3 different qualities of video. The results of our system are conclusive of its high performance." }, "2105.10900.pdf": { "id": "2105.10900v1", "title": "Modeling Collective Anticipation and Response on Wikipedia", "year": "2021", "authors": [ "Ryota Kobayashi", "Patrick Gildersleve", "Takeaki Uno", "Renaud Lambiotte" ], "categories": "cs.SI, cs.CY, physics.soc-ph", "pdf_url": "https://arxiv.org/pdf/2105.10900v1", "abstract": "The dynamics of popularity in online media are driven by a combination of endogenous spreading mechanisms and response to exogenous shocks including news and events. However, little is known about the dependence of temporal patterns of popularity on event-related information, e.g. which types of events trigger long-lasting activity. Here we propose a simple model that describes the dynamics around peaks of popularity by incorporating key features, i.e., the anticipatory growth and the decay of collective attention together with circadian rhythms. The proposed model allows us to develop a new method for predicting the future page view activity and for clustering time series. To validate our methodology, we collect a corpus of page view data from Wikipedia associated to a range of planned events, that are events which we know in advance will have a fixed date in the future, such as elections and sport events. Our methodology is superior to existing models in both prediction and clustering tasks. Furthermore, restricting to Wikipedia pages associated to association football, we observe that the specific realization of the event, in our case which team wins a match or the type of the match, has a significant effect on the response dynamics after the event. Our work demonstrates the importance of appropriately modeling all phases of collective attention, as well as the connection between temporal patterns of attention and characteristic underlying information of the events they represent." }, "2105.12196.pdf": { "id": "2105.12196v1", "title": "From Motor Control to Team Play in Simulated Humanoid Football", "year": "2021", "authors": [ "Siqi Liu", "Guy Lever", "Zhe Wang", "Josh Merel", "S. M. Ali Eslami", "Daniel Hennes" ], "categories": "cs.AI, cs.MA, cs.NE, cs.RO", "pdf_url": "https://arxiv.org/pdf/2105.12196v1", "abstract": "Intelligent behaviour in the physical world exhibits structure at multiple spatial and temporal scales. Although movements are ultimately executed at the level of instantaneous muscle tensions or joint torques, they must be selected to serve goals defined on much longer timescales, and in terms of relations that extend far beyond the body itself, ultimately involving coordination with other agents. Recent research in artificial intelligence has shown the promise of learning-based approaches to the respective problems of complex movement, longer-term planning and multi-agent coordination. However, there is limited research aimed at their integration. We study this problem by training teams of physically simulated humanoid avatars to play football in a realistic virtual environment. We develop a method that combines imitation learning, single- and multi-agent reinforcement learning and population-based training, and makes use of transferable representations of behaviour for decision making at different levels of abstraction. In a sequence of stages, players first learn to control a fully articulated body to perform realistic, human-like movements such as running and turning; they then acquire mid-level football skills such as dribbling and shooting; finally, they develop awareness of others and play as a team, bridging the gap between low-level motor control at a timescale of milliseconds, and coordinated goal-directed behaviour as a team at the timescale of tens of seconds. We investigate the emergence of behaviours at different levels of abstraction, as well as the representations that underlie these behaviours using several analysis techniques, including statistics from real-world sports analytics. Our work constitutes a complete demonstration of integrated decision-making at multiple scales in a physically embodied multi-agent setting. See project video at https://youtu.be/KHMwq9pv7mg." }, "2105.13778.pdf": { "id": "2105.13778v1", "title": "\"Why Would I Trust Your Numbers?\" On the Explainability of Expected Values in Soccer", "year": "2021", "authors": [ "Jan Van Haaren" ], "categories": "cs.LG, stat.AP", "pdf_url": "https://arxiv.org/pdf/2105.13778v1", "abstract": "In recent years, many different approaches have been proposed to quantify the performances of soccer players. Since player performances are challenging to quantify directly due to the low-scoring nature of soccer, most approaches estimate the expected impact of the players' on-the-ball actions on the scoreline. While effective, these approaches are yet to be widely embraced by soccer practitioners. The soccer analytics community has primarily focused on improving the accuracy of the models, while the explainability of the produced metrics is often much more important to practitioners. To help bridge the gap between scientists and practitioners, we introduce an explainable Generalized Additive Model that estimates the expected value for shots. Unlike existing models, our model leverages features corresponding to widespread soccer concepts. To this end, we represent the locations of shots by fuzzily assigning the shots to designated zones on the pitch that practitioners are familiar with. Our experimental evaluation shows that our model is as accurate as existing models, while being easier to explain to soccer practitioners." }, "2106.00173.pdf": { "id": "2106.00173v1", "title": "Enhancing Trajectory Prediction using Sparse Outputs: Application to Team Sports", "year": "2021", "authors": [ "Brandon Victor", "Aiden Nibali", "Zhen He", "David L. Carey" ], "categories": "cs.LG, stat.ML", "pdf_url": "https://arxiv.org/pdf/2106.00173v1", "abstract": "Sophisticated trajectory prediction models that effectively mimic team dynamics have many potential uses for sports coaches, broadcasters and spectators. However, through experiments on soccer data we found that it can be surprisingly challenging to train a deep learning model for player trajectory prediction which outperforms linear extrapolation on average distance between predicted and true future trajectories. We propose and test a novel method for improving training by predicting a sparse trajectory and interpolating using constant acceleration, which improves performance for several models. This interpolation can also be used on models that aren't trained with sparse outputs, and we find that this consistently improves performance for all tested models. Additionally, we find that the accuracy of predicted trajectories for a subset of players can be improved by conditioning on the full trajectories of the other players, and that this is further improved when combined with sparse predictions. We also propose a novel architecture using graph networks and multi-head attention (GraN-MA) which achieves better performance than other tested state-of-the-art models on our dataset and is trivially adapted for both sparse trajectories and full-trajectory conditioned trajectory prediction." }, "2106.00359.pdf": { "id": "2106.00359v1", "title": "Learning Football Body-Orientation as a Matter of Classification", "year": "2021", "authors": [ "Adri\u00e0 Arbu\u00e9s-Sang\u00fcesa", "Adri\u00e1n Mart\u00edn", "Paulino Granero", "Coloma Ballester", "Gloria Haro" ], "categories": "cs.LG, cs.CV, eess.IV", "pdf_url": "https://arxiv.org/pdf/2106.00359v1", "abstract": "Orientation is a crucial skill for football players that becomes a differential factor in a large set of events, especially the ones involving passes. However, existing orientation estimation methods, which are based on computer-vision techniques, still have a lot of room for improvement. To the best of our knowledge, this article presents the first deep learning model for estimating orientation directly from video footage. By approaching this challenge as a classification problem where classes correspond to orientation bins, and by introducing a cyclic loss function, a well-known convolutional network is refined to provide player orientation data. The model is trained by using ground-truth orientation data obtained from wearable EPTS devices, which are individually compensated with respect to the perceived orientation in the current frame. The obtained results outperform previous methods; in particular, the absolute median error is less than 12 degrees per player. An ablation study is included in order to show the potential generalization to any kind of football video footage." }, "2106.01786.pdf": { "id": "2106.01786v1", "title": "What Happened Next? Using Deep Learning to Value Defensive Actions in Football Event-Data", "year": "2021", "authors": [ "Charbel Merhej", "Ryan Beal", "Sarvapali Ramchurn", "Tim Matthews" ], "categories": "cs.AI", "pdf_url": "https://arxiv.org/pdf/2106.01786v1", "abstract": "Objectively quantifying the value of player actions in football (soccer) is a challenging problem. To date, studies in football analytics have mainly focused on the attacking side of the game, while there has been less work on event-driven metrics for valuing defensive actions (e.g., tackles and interceptions). Therefore in this paper, we use deep learning techniques to define a novel metric that values such defensive actions by studying the threat of passages of play that preceded them. By doing so, we are able to value defensive actions based on what they prevented from happening in the game. Our Defensive Action Expected Threat (DAxT) model has been validated using real-world event-data from the 2017/2018 and 2018/2019 English Premier League seasons, and we combine our model outputs with additional features to derive an overall rating of defensive ability for players. Overall, we find that our model is able to predict the impact of defensive actions allowing us to better value defenders using event-data." }, "2106.02195.pdf": { "id": "2106.02195v2", "title": "Celebrating Diversity in Shared Multi-Agent Reinforcement Learning", "year": "2021", "authors": [ "Chenghao Li", "Tonghan Wang", "Chengjie Wu", "Qianchuan Zhao", "Jun Yang", "Chongjie Zhang" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2106.02195v2", "abstract": "Recently, deep multi-agent reinforcement learning (MARL) has shown the promise to solve complex cooperative tasks. Its success is partly because of parameter sharing among agents. However, such sharing may lead agents to behave similarly and limit their coordination capacity. In this paper, we aim to introduce diversity in both optimization and representation of shared multi-agent reinforcement learning. Specifically, we propose an information-theoretical regularization to maximize the mutual information between agents' identities and their trajectories, encouraging extensive exploration and diverse individualized behaviors. In representation, we incorporate agent-specific modules in the shared neural network architecture, which are regularized by L1-norm to promote learning sharing among agents while keeping necessary diversity. Empirical results show that our method achieves state-of-the-art performance on Google Research Football and super hard StarCraft II micromanagement tasks." }, "2106.04219.pdf": { "id": "2106.04219v1", "title": "Time-series Imputation of Temporally-occluded Multiagent Trajectories", "year": "2021", "authors": [ "Shayegan Omidshafiei", "Daniel Hennes", "Marta Garnelo", "Eugene Tarassov", "Zhe Wang", "Romuald Elie" ], "categories": "cs.LG, cs.AI, cs.MA", "pdf_url": "https://arxiv.org/pdf/2106.04219v1", "abstract": "In multiagent environments, several decision-making individuals interact while adhering to the dynamics constraints imposed by the environment. These interactions, combined with the potential stochasticity of the agents' decision-making processes, make such systems complex and interesting to study from a dynamical perspective. Significant research has been conducted on learning models for forward-direction estimation of agent behaviors, for example, pedestrian predictions used for collision-avoidance in self-driving cars. However, in many settings, only sporadic observations of agents may be available in a given trajectory sequence. For instance, in football, subsets of players may come in and out of view of broadcast video footage, while unobserved players continue to interact off-screen. In this paper, we study the problem of multiagent time-series imputation, where available past and future observations of subsets of agents are used to estimate missing observations for other agents. Our approach, called the Graph Imputer, uses forward- and backward-information in combination with graph networks and variational autoencoders to enable learning of a distribution of imputed trajectories. We evaluate our approach on a dataset of football matches, using a projective camera module to train and evaluate our model for the off-screen player state estimation setting. We illustrate that our method outperforms several state-of-the-art approaches, including those hand-crafted for football." }, "2106.05799.pdf": { "id": "2106.05799v1", "title": "Hybrid Machine Learning Forecasts for the UEFA EURO 2020", "year": "2021", "authors": [ "Andreas Groll", "Lars Magnus Hvattum", "Christophe Ley", "Franziska Popp", "Gunther Schauberger", "Hans Van Eetvelde" ], "categories": "cs.LG, stat.AP", "pdf_url": "https://arxiv.org/pdf/2106.05799v1", "abstract": "Three state-of-the-art statistical ranking methods for forecasting football matches are combined with several other predictors in a hybrid machine learning model. Namely an ability estimate for every team based on historic matches; an ability estimate for every team based on bookmaker consensus; average plus-minus player ratings based on their individual performances in their home clubs and national teams; and further team covariates (e.g., market value, team structure) and country-specific socio-economic factors (population, GDP). The proposed combined approach is used for learning the number of goals scored in the matches from the four previous UEFA EUROs 2004-2016 and then applied to current information to forecast the upcoming UEFA EURO 2020. Based on the resulting estimates, the tournament is simulated repeatedly and winning probabilities are obtained for all teams. A random forest model favors the current World Champion France with a winning probability of 14.8% before England (13.5%) and Spain (12.3%). Additionally, we provide survival probabilities for all teams and at all tournament stages." }, "2106.12895.pdf": { "id": "2106.12895v1", "title": "rSoccer: A Framework for Studying Reinforcement Learning in Small and Very Small Size Robot Soccer", "year": "2021", "authors": [ "Felipe B. Martins", "Mateus G. Machado", "Hansenclever F. Bassani", "Pedro H. M. Braga", "Edna S. Barros" ], "categories": "cs.LG, cs.AI, cs.RO", "pdf_url": "https://arxiv.org/pdf/2106.12895v1", "abstract": "Reinforcement learning is an active research area with a vast number of applications in robotics, and the RoboCup competition is an interesting environment for studying and evaluating reinforcement learning methods. A known difficulty in applying reinforcement learning to robotics is the high number of experience samples required, being the use of simulated environments for training the agents followed by transfer learning to real-world (sim-to-real) a viable path. This article introduces an open-source simulator for the IEEE Very Small Size Soccer and the Small Size League optimized for reinforcement learning experiments. We also propose a framework for creating OpenAI Gym environments with a set of benchmarks tasks for evaluating single-agent and multi-agent robot soccer skills. We then demonstrate the learning capabilities of two state-of-the-art reinforcement learning methods as well as their limitations in certain scenarios introduced in this framework. We believe this will make it easier for more teams to compete in these categories using end-to-end reinforcement learning approaches and further develop this research area." }, "2106.14345.pdf": { "id": "2106.14345v2", "title": "More on verification of probability forecasts for football outcomes: score decompositions, reliability, and discrimination analyses", "year": "2021", "authors": [ "Jean-Louis Foulley" ], "categories": "stat.AP", "pdf_url": "https://arxiv.org/pdf/2106.14345v2", "abstract": "Forecast of football outcomes in terms of Home Win, Draw and Away Win relies largely on ex ante probability elicitation of these events and ex post verification of them via computation of probability scoring rules (Brier, Ranked Probability, Logarithmic, Zero-One scores). Usually, appraisal of the quality of forecasting procedures is restricted to reporting mean score values. The purpose of this article is to propose additional tools of verification, such as score decompositions into several components of special interest. Graphical and numerical diagnoses of reliability and discrimination and kindred statistical methods are presented using different techniques of binning (fixed thresholds, quantiles, logistic and iso regression). These procedures are illustrated on probability forecasts for the outcomes of the UEFA Champions League (C1) at the end of the group stage based on typical Poisson regression models with reasonably good results in terms of reliability as compared to those obtained from bookmaker odds and whatever the technique used. Links with research in machine learning and different areas of application (meteorology, medicine) are discussed." }, "2106.14447.pdf": { "id": "2106.14447v1", "title": "Feature Combination Meets Attention: Baidu Soccer Embeddings and Transformer based Temporal Detection", "year": "2021", "authors": [ "Xin Zhou", "Le Kang", "Zhiyu Cheng", "Bo He", "Jingyu Xin" ], "categories": "cs.CV, cs.AI, cs.LG", "pdf_url": "https://arxiv.org/pdf/2106.14447v1", "abstract": "With rapidly evolving internet technologies and emerging tools, sports related videos generated online are increasing at an unprecedentedly fast pace. To automate sports video editing/highlight generation process, a key task is to precisely recognize and locate the events in the long untrimmed videos. In this tech report, we present a two-stage paradigm to detect what and when events happen in soccer broadcast videos. Specifically, we fine-tune multiple action recognition models on soccer data to extract high-level semantic features, and design a transformer based temporal detection module to locate the target events. This approach achieved the state-of-the-art performance in both two tasks, i.e., action spotting and replay grounding, in the SoccerNet-v2 Challenge, under CVPR 2021 ActivityNet workshop. Our soccer embedding features are released at https://github.com/baidu-research/vidpress-sports. By sharing these features with the broader community, we hope to accelerate the research into soccer video understanding." }, "2106.15444.pdf": { "id": "2106.15444v1", "title": "Coach2vec: autoencoding the playing style of soccer coaches", "year": "2021", "authors": [ "Paolo Cintia", "Luca Pappalardo" ], "categories": "cs.AI", "pdf_url": "https://arxiv.org/pdf/2106.15444v1", "abstract": "Capturing the playing style of professional soccer coaches is a complex, and yet barely explored, task in sports analytics. Nowadays, the availability of digital data describing every relevant spatio-temporal aspect of soccer matches, allows for capturing and analyzing the playing style of players, teams, and coaches in an automatic way. In this paper, we present coach2vec, a workflow to capture the playing style of professional coaches using match event streams and artificial intelligence. Coach2vec extracts ball possessions from each match, clusters them based on their similarity, and reconstructs the typical ball possessions of coaches. Then, it uses an autoencoder, a type of artificial neural network, to obtain a concise representation (encoding) of the playing style of each coach. Our experiments, conducted on soccer-logs describing the last four seasons of the Italian first division, reveal interesting similarities between prominent coaches, paving the road to the simulation of playing styles and the quantitative comparison of professional coaches." }, "2107.05714.pdf": { "id": "2107.05714v2", "title": "The Voronoi Diagram in Soccer: a theoretical study to measure dominance space", "year": "2021", "authors": [ "Costas J. Efthimiou" ], "categories": "physics.soc-ph, physics.pop-ph", "pdf_url": "https://arxiv.org/pdf/2107.05714v2", "abstract": "In this article, I argue that the concept of the standard Voronoi diagram is incorrect for the calculation of dominance area in soccer. The correct region that should replace the polygonal Voronoi region controlled by a player emerges naturally by studying the motion of the players. It turns out that the widely-used standard Voronoi diagram is true only when all 22 players have the same speed (with vanishing speed being a special case) and it should be replaced by other diagrams whose regions have more complicated boundaries. The unfortunate error happened when the mathematical concept of the Voronoi diagram that has an implicit vanishing speed for its points was transferred to soccer without looking at the kinematical issues involved in the game of soccer. For this reason, as a byproduct of this work, I argue for the promotion of \"soccerdynamics\", model building for soccer by looking at the underlying mechanisms and not just adopting mathematical ideas only for their appeal and beauty." }, "2107.08732.pdf": { "id": "2107.08732v2", "title": "Assessing competitive balance in the English Premier League for over forty seasons using a stochastic block model", "year": "2021", "authors": [ "Francesca Basini", "Vasiliki Tsouli", "Ioannis Ntzoufras", "Nial Friel" ], "categories": "stat.AP", "pdf_url": "https://arxiv.org/pdf/2107.08732v2", "abstract": "Competitive balance is the subject of much interest in the sports analytics literature and beyond. In this paper, we develop a statistical network model based on an extension of the stochastic block model to assess the balance between teams in a league. Here we represent the outcome of all matches in a football season as a dense network with nodes identified by teams and categorical edges representing the outcome of each game as a win, draw or a loss. The main focus and motivation for this paper is to provide a statistical framework to assess the issue of competitive balance in the context of the English First Division / Premier League over more than 40 seasons. The Premier League is arguably one of the most popular leagues in the world, in terms of its global reach and the revenue which it generates. Therefore it is of wide interest to assess its competitiveness. Our analysis provides evidence suggesting a structural change around the early 2000's from a reasonably balanced league to a two-tier league." }, "2108.03498.pdf": { "id": "2108.03498v1", "title": "Kinematics clustering enables head impact subtyping for better traumatic brain injury prediction", "year": "2021", "authors": [ "Xianghao Zhan", "Yiheng Li", "Yuzhe Liu", "Nicholas J. Cecchi", "Olivier Gevaert", "Michael M. Zeineh" ], "categories": "stat.AP, cs.LG, q-bio.QM, q-bio.TO", "pdf_url": "https://arxiv.org/pdf/2108.03498v1", "abstract": "Traumatic brain injury can be caused by various types of head impacts. However, due to different kinematic characteristics, many brain injury risk estimation models are not generalizable across the variety of impacts that humans may sustain. The current definitions of head impact subtypes are based on impact sources (e.g., football, traffic accident), which may not reflect the intrinsic kinematic similarities of impacts across the impact sources. To investigate the potential new definitions of impact subtypes based on kinematics, 3,161 head impacts from various sources including simulation, college football, mixed martial arts, and car racing were collected. We applied the K-means clustering to cluster the impacts on 16 standardized temporal features from head rotation kinematics. Then, we developed subtype-specific ridge regression models for cumulative strain damage (using the threshold of 15%), which significantly improved the estimation accuracy compared with the baseline method which mixed impacts from different sources and developed one model (R^2 from 0.7 to 0.9). To investigate the effect of kinematic features, we presented the top three critical features (maximum resultant angular acceleration, maximum angular acceleration along the z-axis, maximum linear acceleration along the y-axis) based on regression accuracy and used logistic regression to find the critical points for each feature that partitioned the subtypes. This study enables researchers to define head impact subtypes in a data-driven manner, which leads to more generalizable brain injury risk estimation." }, "2108.05343.pdf": { "id": "2108.05343v1", "title": "Effect of Key Match Events on Football Passmaps", "year": "2021", "authors": [ "Vid Stropnik", "Vuk \u0110uranovi\u0107", "Maru\u0161a Ora\u017eem" ], "categories": "cs.SI", "pdf_url": "https://arxiv.org/pdf/2108.05343v1", "abstract": "A team of association football players may be envisioned as a directed network with player nodes and weighted pass edges. Such a simplistic representation of an otherwise complex structure yields several benefits, but also permits the application of well-established network analysis approaches, to gain insights into the match flow. The authors of this work propose the application of these, including centrality measures, community detection methods, and the analysis of induced subgraphs, to analyze the potential effects certain match events have on the team. In addition to reporting on their findings, they emphasize their work's contribution of over 9000 novel football network representations to the Network Analysis community." }, "2108.05796.pdf": { "id": "2108.05796v1", "title": "Goal scoring in Premier League with Poisson regression", "year": "2021", "authors": [ "Cuong Pham", "Tung Le" ], "categories": "stat.AP, cs.LG", "pdf_url": "https://arxiv.org/pdf/2108.05796v1", "abstract": "Premier League is known as one of the most competitive football league in the world, hence there are many goals are scored here every match. Which are the factors that affect to the number of goal scored in each match? We use Poisson regression to find out the relation between many factors as shots on target, corners, red cards, to the goals home team can score in their match." }, "2108.09478.pdf": { "id": "2108.09478v1", "title": "MimicBot: Combining Imitation and Reinforcement Learning to win in Bot Bowl", "year": "2021", "authors": [ "Nicola Pezzotti" ], "categories": "cs.AI, cs.LG", "pdf_url": "https://arxiv.org/pdf/2108.09478v1", "abstract": "This paper describe an hybrid agent trained to play in Fantasy Football AI which participated in the Bot Bowl III competition. The agent, MimicBot, is implemented using a specifically designed deep policy network and trained using a combination of imitation and reinforcement learning. Previous attempts in using a reinforcement learning approach in such context failed for a number of reasons, e.g. due to the intrinsic randomness in the environment and the large and uneven number of actions available, with a curriculum learning approach failing to consistently beat a randomly paying agent. Currently no machine learning approach can beat a scripted bot which makes use of the domain knowledge on the game. Our solution, thanks to an imitation learning and a hybrid decision-making process, consistently beat such scripted agents. Moreover we shed lights on how to more efficiently train in a reinforcement learning setting while drastically increasing sample efficiency. MimicBot is the winner of the Bot Bowl III competition, and it is currently the state-of-the-art solution." }, "2109.04720.pdf": { "id": "2109.04720v1", "title": "6MapNet: Representing soccer players from tracking data by a triplet network", "year": "2021", "authors": [ "Hyunsung Kim", "Jihun Kim", "Dongwook Chung", "Jonghyun Lee", "Jinsung Yoon", "Sang-Ki Ko" ], "categories": "cs.LG, cs.AI", "pdf_url": "https://arxiv.org/pdf/2109.04720v1", "abstract": "Although the values of individual soccer players have become astronomical, subjective judgments still play a big part in the player analysis. Recently, there have been new attempts to quantitatively grasp players' styles using video-based event stream data. However, they have some limitations in scalability due to high annotation costs and sparsity of event stream data. In this paper, we build a triplet network named 6MapNet that can effectively capture the movement styles of players using in-game GPS data. Without any annotation of soccer-specific actions, we use players' locations and velocities to generate two types of heatmaps. Our subnetworks then map these heatmap pairs into feature vectors whose similarity corresponds to the actual similarity of playing styles. The experimental results show that players can be accurately identified with only a small number of matches by our method." }, "2109.06625.pdf": { "id": "2109.06625v1", "title": "Towards optimized actions in critical situations of soccer games with deep reinforcement learning", "year": "2021", "authors": [ "Pegah Rahimian", "Afshin Oroojlooy", "Laszlo Toka" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2109.06625v1", "abstract": "Soccer is a sparse rewarding game: any smart or careless action in critical situations can change the result of the match. Therefore players, coaches, and scouts are all curious about the best action to be performed in critical situations, such as the times with a high probability of losing ball possession or scoring a goal. This work proposes a new state representation for the soccer game and a batch reinforcement learning to train a smart policy network. This network gets the contextual information of the situation and proposes the optimal action to maximize the expected goal for the team. We performed extensive numerical experiments on the soccer logs made by InStat for 104 European soccer matches. The results show that in all 104 games, the optimized policy obtains higher rewards than its counterpart in the behavior policy. Besides, our framework learns policies that are close to the expected behavior in the real world. For instance, in the optimized policy, we observe that some actions such as foul, or ball out can be sometimes more rewarding than a shot in specific situations." }, "2110.04321.pdf": { "id": "2110.04321v1", "title": "Computing an Optimal Pitching Strategy in a Baseball At-Bat", "year": "2021", "authors": [ "Connor Douglas", "Everett Witt", "Mia Bendy", "Yevgeniy Vorobeychik" ], "categories": "cs.GT, cs.AI, cs.LG, cs.MA", "pdf_url": "https://arxiv.org/pdf/2110.04321v1", "abstract": "The field of quantitative analytics has transformed the world of sports over the last decade. To date, these analytic approaches are statistical at their core, characterizing what is and what was, while using this information to drive decisions about what to do in the future. However, as we often view team sports, such as soccer, hockey, and baseball, as pairwise win-lose encounters, it seems natural to model these as zero-sum games. We propose such a model for one important class of sports encounters: a baseball at-bat, which is a matchup between a pitcher and a batter. Specifically, we propose a novel model of this encounter as a zero-sum stochastic game, in which the goal of the batter is to get on base, an outcome the pitcher aims to prevent. The value of this game is the on-base percentage (i.e., the probability that the batter gets on base). In principle, this stochastic game can be solved using classical approaches. The main technical challenges lie in predicting the distribution of pitch locations as a function of pitcher intention, predicting the distribution of outcomes if the batter decides to swing at a pitch, and characterizing the level of patience of a particular batter. We address these challenges by proposing novel pitcher and batter representations as well as a novel deep neural network architecture for outcome prediction. Our experiments using Kaggle data from the 2015 to 2018 Major League Baseball seasons demonstrate the efficacy of the proposed approach." }, "2110.05362.pdf": { "id": "2110.05362v1", "title": "Focus on what matters: Applying Discourse Coherence Theory to Cross Document Coreference", "year": "2021", "authors": [ "William Held", "Dan Iter", "Dan Jurafsky" ], "categories": "cs.CL, cs.LG", "pdf_url": "https://arxiv.org/pdf/2110.05362v1", "abstract": "Performing event and entity coreference resolution across documents vastly increases the number of candidate mentions, making it intractable to do the full $n^2$ pairwise comparisons. Existing approaches simplify by considering coreference only within document clusters, but this fails to handle inter-cluster coreference, common in many applications. As a result cross-document coreference algorithms are rarely applied to downstream tasks. We draw on an insight from discourse coherence theory: potential coreferences are constrained by the reader's discourse focus. We model the entities/events in a reader's focus as a neighborhood within a learned latent embedding space which minimizes the distance between mentions and the centroids of their gold coreference clusters. We then use these neighborhoods to sample only hard negatives to train a fine-grained classifier on mention pairs and their local discourse features. Our approach achieves state-of-the-art results for both events and entities on the ECB+, Gun Violence, Football Coreference, and Cross-Domain Cross-Document Coreference corpora. Furthermore, training on multiple corpora improves average performance across all datasets by 17.2 F1 points, leading to a robust coreference resolution model for use in downstream tasks where link distribution is unknown." }, "2110.07953.pdf": { "id": "2110.07953v2", "title": "Attention-based Estimation and Prediction of Human Intent to augment Haptic Glove aided Control of Robotic Hand", "year": "2021", "authors": [ "Muneeb Ahmed", "Rajesh Kumar", "Qaim Abbas", "Brejesh Lall", "Arzad A. Kherani", "Sudipto Mukherjee" ], "categories": "cs.RO, cs.AI", "pdf_url": "https://arxiv.org/pdf/2110.07953v2", "abstract": "The letter focuses on Haptic Glove (HG) based control of a Robotic Hand (RH) executing in-hand manipulation of certain objects of interest. The high dimensional motion signals in HG and RH possess intrinsic variability of kinematics resulting in difficulty to establish a direct mapping of the motion signals from HG onto the RH. An estimation mechanism is proposed to quantify the motion signal acquired from the human controller in relation to the intended goal pose of the object being held by the robotic hand. A control algorithm is presented to transform the synthesized intent at the RH and allow relocation of the object to the expected goal pose. The lag in synthesis of the intent in the presence of communication delay leads to a requirement of predicting the estimated intent. We leverage an attention-based convolutional neural network encoder to predict the trajectory of intent for a certain lookahead to compensate for the delays. The proposed methodology is evaluated across objects of different shapes, mass, and materials. We present a comparative performance of the estimation and prediction mechanisms on 5G-driven real-world robotic setup against benchmark methodologies. The test-MSE in prediction of human intent is reported to yield ~ 97.3 -98.7% improvement of accuracy in comparison to LSTM-based benchmark" }, "2110.08169.pdf": { "id": "2110.08169v2", "title": "Containerized Distributed Value-Based Multi-Agent Reinforcement Learning", "year": "2021", "authors": [ "Siyang Wu", "Tonghan Wang", "Chenghao Li", "Yang Hu", "Chongjie Zhang" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2110.08169v2", "abstract": "Multi-agent reinforcement learning tasks put a high demand on the volume of training samples. Different from its single-agent counterpart, distributed value-based multi-agent reinforcement learning faces the unique challenges of demanding data transfer, inter-process communication management, and high requirement of exploration. We propose a containerized learning framework to solve these problems. We pack several environment instances, a local learner and buffer, and a carefully designed multi-queue manager which avoids blocking into a container. Local policies of each container are encouraged to be as diverse as possible, and only trajectories with highest priority are sent to a global learner. In this way, we achieve a scalable, time-efficient, and diverse distributed MARL learning framework with high system throughput. To own knowledge, our method is the first to solve the challenging Google Research Football full game $5\\_v\\_5$. On the StarCraft II micromanagement benchmark, our method gets $4$-$18\\times$ better results compared to state-of-the-art non-distributed MARL algorithms." }, "2111.02874.pdf": { "id": "2111.02874v1", "title": "Deep Artificial Intelligence for Fantasy Football Language Understanding", "year": "2021", "authors": [ "Aaron Baughman", "Micah Forester", "Jeff Powell", "Eduardo Morales", "Shaun McPartlin", "Daniel Bohm" ], "categories": "cs.AI, cs.LG", "pdf_url": "https://arxiv.org/pdf/2111.02874v1", "abstract": "Fantasy sports allow fans to manage a team of their favorite athletes and compete with friends. The fantasy platform aligns the real-world statistical performance of athletes to fantasy scoring and has steadily risen in popularity to an estimated 9.1 million players per month with 4.4 billion player card views on the ESPN Fantasy Football platform from 2018-2019. In parallel, the sports media community produces news stories, blogs, forum posts, tweets, videos, podcasts and opinion pieces that are both within and outside the context of fantasy sports. However, human fantasy football players can only analyze an average of 3.9 sources of information. Our work discusses the results of a machine learning pipeline to manage an ESPN Fantasy Football team. The use of trained statistical entity detectors and document2vector models applied to over 100,000 news sources and 2.3 million articles, videos and podcasts each day enables the system to comprehend natural language with an analogy test accuracy of 100% and keyword test accuracy of 80%. Deep learning feedforward neural networks provide player classifications such as if a player will be a bust, boom, play with a hidden injury or play meaningful touches with a cumulative 72% accuracy. Finally, a multiple regression ensemble uses the deep learning output and ESPN projection data to provide a point projection for each of the top 500+ fantasy football players in 2018. The point projection maintained a RMSE of 6.78 points. The best fit probability density function from a set of 24 is selected to visualize score spreads. Within the first 6 weeks of the product launch, the total number of users spent a cumulative time of over 4.6 years viewing our AI insights. The training data for our models was provided by a 2015 to 2016 web archive from Webhose, ESPN statistics, and Rotowire injury reports. We used 2017 fantasy football data as a test set." }, "2111.03435.pdf": { "id": "2111.03435v1", "title": "On the role of tournament design in sporting success: A study of the North, Central American and Caribbean qualification for the 2022 FIFA World Cup", "year": "2021", "authors": [ "L\u00e1szl\u00f3 Csat\u00f3" ], "categories": "physics.soc-ph", "pdf_url": "https://arxiv.org/pdf/2111.03435v1", "abstract": "Playing in the FIFA World Cup finals is an ambition shared by several nations. Since, besides luck and skill, the probability of qualification depends on the design of the qualifiers, the study of these competitions forms an integral part of sports analytics. The Confederation of North, Central America and Caribbean Association Football (CONCACAF) announced a novel qualifying format for the 2022 FIFA World Cup in July 2019. However, the COVID-19 pandemic forced the organisers to return to a more traditional structure. The present chapter analyses how this reform impacted the chances of the national teams to qualify. It is found that the probability of participating in the FIFA World Cup finals can change by more than 5 percentage points under the assumption of fixed strengths for the teams. The idea behind the original design, to divide the contestants into two distinct sets, is worth considering due to the increased competitiveness of the matches played by the strongest and the weakest teams. We recommend mitigating the sharp nonlinearity caused by the seeding policy via a probabilistic rule to the analogy of the NBA draft lottery system." }, "2112.08409.pdf": { "id": "2112.08409v1", "title": "Quantum Model Learning Agent: characterisation of quantum systems through machine learning", "year": "2021", "authors": [ "Brian Flynn", "Antonio Andreas Gentile", "Nathan Wiebe", "Raffaele Santagati", "Anthony Laing" ], "categories": "quant-ph, cs.LG", "pdf_url": "https://arxiv.org/pdf/2112.08409v1", "abstract": "Accurate models of real quantum systems are important for investigating their behaviour, yet are difficult to distill empirically. Here, we report an algorithm -- the Quantum Model Learning Agent (QMLA) -- to reverse engineer Hamiltonian descriptions of a target system. We test the performance of QMLA on a number of simulated experiments, demonstrating several mechanisms for the design of candidate Hamiltonian models and simultaneously entertaining numerous hypotheses about the nature of the physical interactions governing the system under study. QMLA is shown to identify the true model in the majority of instances, when provided with limited a priori information, and control of the experimental setup. Our protocol can explore Ising, Heisenberg and Hubbard families of models in parallel, reliably identifying the family which best describes the system dynamics. We demonstrate QMLA operating on large model spaces by incorporating a genetic algorithm to formulate new hypothetical models. The selection of models whose features propagate to the next generation is based upon an objective function inspired by the Elo rating scheme, typically used to rate competitors in games such as chess and football. In all instances, our protocol finds models that exhibit $F_1$-score $\\geq 0.88$ when compared with the true model, and it precisely identifies the true model in 72% of cases, whilst exploring a space of over $250,000$ potential models. By testing which interactions actually occur in the target system, QMLA is a viable tool for both the exploration of fundamental physics and the characterisation and calibration of quantum devices." }, "2112.13001.pdf": { "id": "2112.13001v3", "title": "Forecasting number of corner kicks taken in association football using compound Poisson distribution", "year": "2021", "authors": [ "Stan Yip", "Yinghong Zou", "Ronald Tsz Hin Hung", "Ka Fai Cedric Yiu" ], "categories": "stat.AP", "pdf_url": "https://arxiv.org/pdf/2112.13001v3", "abstract": "This article presents a holistic compound Poisson regression model framework to forecast number of corner kicks taken in association football. Corner kick taken events are often decisive in the match outcome and inherently arrive in batch with serial clustering pattern. Providing parameter estimates with intuitive interpretation, a class of compound Poisson regression including a Bayesian implementation of geometric-Poisson distribution is introduced. With a varying shape parameter, the corner counts serial correlation between matches is handled naturally within the Bayesian model. In this study, information elicited from cross-market betting odds was used to improve the model predictability. Margin application methods to adjust market inefficiency in raw odds are also discussed." }, "2112.13668.pdf": { "id": "2112.13668v1", "title": "Optimizing decision making for soccer line-up by a quantum annealer", "year": "2021", "authors": [ "Aitzol Iturrospe" ], "categories": "quant-ph, math.OC", "pdf_url": "https://arxiv.org/pdf/2112.13668v1", "abstract": "In this paper an application of D-Wave Systems quantum annealer for optimizing initial line-up of a soccer team is presented. Players and their playing position are selected to maximize the sum of players ratings for both a 4-3-3 attack formation and a 4-2-3-1 medium defensive formation. The problem is stated as a binary quadratic model (BQM) and it is solved in a D-Wave Leap Hybrid Solver. Results are presented and compared with already published results obtained with a classical solver." }, "2201.00691.pdf": { "id": "2201.00691v1", "title": "FIFA ranking: Evaluation and path forward", "year": "2021", "authors": [ "Leszek Szczecinski", "Iris-Ioana Roatis" ], "categories": "cs.IR, cs.LG, stat.AP", "pdf_url": "https://arxiv.org/pdf/2201.00691v1", "abstract": "In this work we study the ranking algorithm used by F\u00e9d\u00e9ration Internationale de Football Association (FIFA); we analyze the parameters it currently uses, show the formal probabilistic model from which it can be derived, and optimize the latter. In particular, analyzing the games since the introduction of the algorithm in 2018, we conclude that the game's \"importance\" (as defined by FIFA) used in the algorithm is counterproductive from the point of view of the predictive capability of the algorithm. We also postulate the algorithm to be rooted in the formal modelling principle, where the Davidson model proposed in 1970 seems to be an excellent candidate, preserving the form of the algorithm currently used. The results indicate that the predictive capability of the algorithm is notably improved by using the home-field advantage and the explicit model for the draws in the game. Moderate, but notable improvement may be attained by introducing the weighting of the results with the goal differential, which although not rooted in a formal modelling principle, is compatible with the current algorithm and can be tuned to the characteristics of the football competition." }, "2201.11533.pdf": { "id": "2201.11533v1", "title": "Transfer Portal: Accurately Forecasting the Impact of a Player Transfer in Soccer", "year": "2022", "authors": [ "Daniel Dinsdale", "Joe Gallagher" ], "categories": "cs.LG, stat.AP, stat.ML", "pdf_url": "https://arxiv.org/pdf/2201.11533v1", "abstract": "One of the most important and challenging problems in football is predicting future player performance when transferred to another club within and between different leagues. In addition to being the most valuable prediction a team makes, it is also the most complex analytics task to perform as it needs to take into consideration: a) differences in playing style between the player's current team and target team, b) differences in style and ability of other players on each team, c) differences in league quality and style, and d) the role the player is desired to play. In this paper, we present a method which addresses these issues and enables us to make accurate predictions of future performance. Our Transfer Portal model utilizes a personalized neural network accounting for both stylistic and ability level input representations for players, teams, and leagues to simulate future player performance at any chosen club. Furthermore, we use a Bayesian updating framework to dynamically modify player and team representations over time which enables us to generate predictions for rising stars with small amounts of data." }, "2202.00804.pdf": { "id": "2202.00804v3", "title": "Automatic event detection in football using tracking data", "year": "2022", "authors": [ "Ferran Vidal-Codina", "Nicolas Evans", "Bahaeddine El Fakir", "Johsan Billingham" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2202.00804v3", "abstract": "One of the main shortcomings of event data in football, which has been extensively used for analytics in the recent years, is that it still requires manual collection, thus limiting its availability to a reduced number of tournaments. In this work, we propose a deterministic decision tree-based algorithm to automatically extract football events using tracking data, which consists of two steps: (1) a possession step that evaluates which player was in possession of the ball at each frame in the tracking data, as well as the distinct player configurations during the time intervals where the ball is not in play to inform set piece detection; (2) an event detection step that combines the changes in ball possession computed in the first step with the laws of football to determine in-game events and set pieces. The automatically generated events are benchmarked against manually annotated events and we show that in most event categories the proposed methodology achieves $+90\\%$ detection rate across different tournaments and tracking data providers. Finally, we demonstrate how the contextual information offered by tracking data can be leveraged to increase the granularity of auto-detected events, and exhibit how the proposed framework may be used to conduct a myriad of data analyses in football." }, "2202.01031.pdf": { "id": "2202.01031v1", "title": "MMSys'22 Grand Challenge on AI-based Video Production for Soccer", "year": "2022", "authors": [ "Cise Midoglu", "Steven A. Hicks", "Vajira Thambawita", "Tomas Kupka", "P\u00e5l Halvorsen" ], "categories": "cs.CV, cs.MM", "pdf_url": "https://arxiv.org/pdf/2202.01031v1", "abstract": "Soccer has a considerable market share of the global sports industry, and the interest in viewing videos from soccer games continues to grow. In this respect, it is important to provide game summaries and highlights of the main game events. However, annotating and producing events and summaries often require expensive equipment and a lot of tedious, cumbersome, manual labor. Therefore, automating the video production pipeline providing fast game highlights at a much lower cost is seen as the \"holy grail\". In this context, recent developments in Artificial Intelligence (AI) technology have shown great potential. Still, state-of-the-art approaches are far from being adequate for practical scenarios that have demanding real-time requirements, as well as strict performance criteria (where at least the detection of official events such as goals and cards must be 100% accurate). In addition, event detection should be thoroughly enhanced by annotation and classification, proper clipping, generating short descriptions, selecting appropriate thumbnails for highlight clips, and finally, combining the event highlights into an overall game summary, similar to what is commonly aired during sports news. Even though the event tagging operation has by far received the most attention, an end-to-end video production pipeline also includes various other operations which serve the overall purpose of automated soccer analysis. This challenge aims to assist the automation of such a production pipeline using AI. In particular, we focus on the enhancement operations that take place after an event has been detected, namely event clipping (Task 1), thumbnail selection (Task 2), and game summarization (Task 3). Challenge website: https://mmsys2022.ie/authors/grand-challenge." }, "2202.11204.pdf": { "id": "2202.11204v4", "title": "Study of Feature Importance for Quantum Machine Learning Models", "year": "2022", "authors": [ "Aaron Baughman", "Kavitha Yogaraj", "Raja Hebbar", "Sudeep Ghosh", "Rukhsan Ul Haq", "Yoshika Chhabra" ], "categories": "quant-ph, cs.LG", "pdf_url": "https://arxiv.org/pdf/2202.11204v4", "abstract": "Predictor importance is a crucial part of data preprocessing pipelines in classical and quantum machine learning (QML). This work presents the first study of its kind in which feature importance for QML models has been explored and contrasted against their classical machine learning (CML) equivalents. We developed a hybrid quantum-classical architecture where QML models are trained and feature importance values are calculated from classical algorithms on a real-world dataset. This architecture has been implemented on ESPN Fantasy Football data using Qiskit statevector simulators and IBM quantum hardware such as the IBMQ Mumbai and IBMQ Montreal systems. Even though we are in the Noisy Intermediate-Scale Quantum (NISQ) era, the physical quantum computing results are promising. To facilitate current quantum scale, we created a data tiering, model aggregation, and novel validation methods. Notably, the feature importance magnitudes from the quantum models had a much higher variation when contrasted to classical models. We can show that equivalent QML and CML models are complementary through diversity measurements. The diversity between QML and CML demonstrates that both approaches can contribute to a solution in different ways. Within this paper we focus on Quantum Support Vector Classifiers (QSVC), Variational Quantum Circuit (VQC), and their classical counterparts. The ESPN and IBM fantasy football Trade Assistant combines advanced statistical analysis with the natural language processing of Watson Discovery to serve up personalized trade recommendations that are fair. Here, player valuation data of each player has been considered and this work can be extended to calculate the feature importance of other QML models such as Quantum Boltzmann machines." }, "2203.02281.pdf": { "id": "2203.02281v2", "title": "A Comprehensive Review of Computer Vision in Sports: Open Issues, Future Trends and Research Directions", "year": "2022", "authors": [ "Banoth Thulasya Naik", "Mohammad Farukh Hashmi", "Neeraj Dhanraj Bokde" ], "categories": "cs.CV, eess.IV", "pdf_url": "https://arxiv.org/pdf/2203.02281v2", "abstract": "Recent developments in video analysis of sports and computer vision techniques have achieved significant improvements to enable a variety of critical operations. To provide enhanced information, such as detailed complex analysis in sports like soccer, basketball, cricket, badminton, etc., studies have focused mainly on computer vision techniques employed to carry out different tasks. This paper presents a comprehensive review of sports video analysis for various applications high-level analysis such as detection and classification of players, tracking player or ball in sports and predicting the trajectories of player or ball, recognizing the teams strategies, classifying various events in sports. The paper further discusses published works in a variety of application-specific tasks related to sports and the present researchers views regarding them. Since there is a wide research scope in sports for deploying computer vision techniques in various sports, some of the publicly available datasets related to a particular sport have been provided. This work reviews a detailed discussion on some of the artificial intelligence(AI)applications in sports vision, GPU-based work stations, and embedded platforms. Finally, this review identifies the research directions, probable challenges, and future trends in the area of visual recognition in sports." }, "2203.06933.pdf": { "id": "2203.06933v3", "title": "Soccer: a quantitative analysis of team resilience and the miracle of Bern", "year": "2022", "authors": [ "Ralph St\u00f6mmer" ], "categories": "stat.AP, physics.pop-ph, physics.soc-ph", "pdf_url": "https://arxiv.org/pdf/2203.06933v3", "abstract": "Resilience is the ability to positively respond to adversity. It has been studied in psychology for several decades, with focus on how individuals overcome traumata or cope with setbacks and obstacles in their professional career. Research on resilience in the sport context is rather new. Activities are based on insights that in highly competitive environments, tiny effects tip the scales. A key question of measuring resilience in sports is what parameters to measure. Here a novel concept is proposed to measure the resilience of soccer teams. The relative frequency of matches is determined, where a soccer team, which is initially trailing by 2 goals, finally succeeds to win the match or at least to reach a draw. The analysis is applied to the last 59 seasons of the German premier soccer league Bundesliga. The empirical data are compared with a theoretical model derived from Poisson distributions. It is shown how leading teams in the premier soccer league differ from the average with respect to resilience, which provides further insights into the hidden secrets of top soccer teams." }, "2204.00003.pdf": { "id": "2204.00003v3", "title": "Ball 3D Localization From A Single Calibrated Image", "year": "2022", "authors": [ "Gabriel Van Zandycke", "Christophe De Vleeschouwer" ], "categories": "cs.CV, eess.IV", "pdf_url": "https://arxiv.org/pdf/2204.00003v3", "abstract": "Ball 3D localization in team sports has various applications including automatic offside detection in soccer, or shot release localization in basketball. Today, this task is either resolved by using expensive multi-views setups, or by restricting the analysis to ballistic trajectories. In this work, we propose to address the task on a single image from a calibrated monocular camera by estimating ball diameter in pixels and use the knowledge of real ball diameter in meters. This approach is suitable for any game situation where the ball is (even partly) visible. To achieve this, we use a small neural network trained on image patches around candidates generated by a conventional ball detector. Besides predicting ball diameter, our network outputs the confidence of having a ball in the image patch. Validations on 3 basketball datasets reveals that our model gives remarkable predictions on ball 3D localization. In addition, through its confidence output, our model improves the detection rate by filtering the candidates produced by the detector. The contributions of this work are (i) the first model to address 3D ball localization on a single image, (ii) an effective method for ball 3D annotation from single calibrated images, (iii) a high quality 3D ball evaluation dataset annotated from a single viewpoint. In addition, the code to reproduce this research is be made freely available at https://github.com/gabriel-vanzandycke/deepsport." }, "2204.02313.pdf": { "id": "2204.02313v1", "title": "Is it worth the effort? Understanding and contextualizing physical metrics in soccer", "year": "2022", "authors": [ "Sergio Llana", "Borja Burriel", "Pau Madrero", "Javier Fern\u00e1ndez" ], "categories": "stat.ML, cs.LG", "pdf_url": "https://arxiv.org/pdf/2204.02313v1", "abstract": "We present a framework that gives a deep insight into the link between physical and technical-tactical aspects of soccer and it allows associating physical performance with value generation thanks to a top-down approach. First, we estimate physical indicators from tracking data. Then, we contextualize each player's run to understand better the purpose and circumstances in which it is done, adding a new dimension to the creation of team and player profiles. Finally, we assess the value-added by off-ball high-intensity runs by linking with a possession-value model. This novel approach allows answering practical questions from very different profiles of practitioners within a soccer club, from analysts, coaches, and scouts to physical coaches and readaptation physiotherapists." }, "2204.02573.pdf": { "id": "2204.02573v1", "title": "Detecting key Soccer match events to create highlights using Computer Vision", "year": "2022", "authors": [ "Narayana Darapaneni", "Prashant Kumar", "Nikhil Malhotra", "Vigneswaran Sundaramurthy", "Abhaya Thakur", "Shivam Chauhan" ], "categories": "cs.CV, cs.AI", "pdf_url": "https://arxiv.org/pdf/2204.02573v1", "abstract": "The research and data science community has been fascinated with the development of automatic systems for the detection of key events in a video. Special attention in this field is given to sports video analytics which could help in identifying key events during a match and help in preparing a strategy for the games going forward. For this paper, we have chosen Football (soccer) as a sport where we would want to create highlights for a given match video, through a computer vision model that aims to identify important events in a Soccer match to create highlights of the match. We built the models based on Faster RCNN and YoloV5 architectures and noticed that for the amount of data we used for training Faster RCNN did better than YoloV5 in detecting the events in the match though it was much slower. Within Faster RCNN using ResNet50 as a base model gave a better class accuracy of 95.5% as compared to 92% with VGG16 as base model completely outperforming YoloV5 for our training dataset. We tested with an original video of size 23 minutes and our model could reduce it to 4:50 minutes of highlights capturing almost all important events in the match." }, "2204.06859.pdf": { "id": "2204.06859v1", "title": "Semi-Supervised Training to Improve Player and Ball Detection in Soccer", "year": "2022", "authors": [ "Renaud Vandeghen", "Anthony Cioppa", "Marc Van Droogenbroeck" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2204.06859v1", "abstract": "Accurate player and ball detection has become increasingly important in recent years for sport analytics. As most state-of-the-art methods rely on training deep learning networks in a supervised fashion, they require huge amounts of annotated data, which are rarely available. In this paper, we present a novel generic semi-supervised method to train a network based on a labeled image dataset by leveraging a large unlabeled dataset of soccer broadcast videos. More precisely, we design a teacher-student approach in which the teacher produces surrogate annotations on the unlabeled data to be used later for training a student which has the same architecture as the teacher. Furthermore, we introduce three training loss parametrizations that allow the student to doubt the predictions of the teacher during training depending on the proposal confidence score. We show that including unlabeled data in the training process allows to substantially improve the performances of the detection network trained only on the labeled data. Finally, we provide a thorough performance study including different proportions of labeled and unlabeled data, and establish the first benchmark on the new SoccerNet-v3 detection task, with an mAP of 52.3%. Our code is available at https://github.com/rvandeghen/SST ." }, "2204.06918.pdf": { "id": "2204.06918v2", "title": "SoccerNet-Tracking: Multiple Object Tracking Dataset and Benchmark in Soccer Videos", "year": "2022", "authors": [ "Anthony Cioppa", "Silvio Giancola", "Adrien Deliege", "Le Kang", "Xin Zhou", "Zhiyu Cheng" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2204.06918v2", "abstract": "Tracking objects in soccer videos is extremely important to gather both player and team statistics, whether it is to estimate the total distance run, the ball possession or the team formation. Video processing can help automating the extraction of those information, without the need of any invasive sensor, hence applicable to any team on any stadium. Yet, the availability of datasets to train learnable models and benchmarks to evaluate methods on a common testbed is very limited. In this work, we propose a novel dataset for multiple object tracking composed of 200 sequences of 30s each, representative of challenging soccer scenarios, and a complete 45-minutes half-time for long-term tracking. The dataset is fully annotated with bounding boxes and tracklet IDs, enabling the training of MOT baselines in the soccer domain and a full benchmarking of those methods on our segregated challenge sets. Our analysis shows that multiple player, referee and ball tracking in soccer videos is far from being solved, with several improvement required in case of fast motion or in scenarios of severe occlusion." }, "2204.11142.pdf": { "id": "2204.11142v1", "title": "Graph Neural Network based Agent in Google Research Football", "year": "2022", "authors": [ "Yizhan Niu", "Jinglong Liu", "Yuhao Shi", "Jiren Zhu" ], "categories": "cs.LG, cs.AI", "pdf_url": "https://arxiv.org/pdf/2204.11142v1", "abstract": "Deep neural networks (DNN) can approximate value functions or policies for reinforcement learning, which makes the reinforcement learning algorithms more powerful. However, some DNNs, such as convolutional neural networks (CNN), cannot extract enough information or take too long to obtain enough features from the inputs under specific circumstances of reinforcement learning. For example, the input data of Google Research Football, a reinforcement learning environment which trains agents to play football, is the small map of players' locations. The information is contained not only in the coordinates of players, but also in the relationships between different players. CNNs can neither extract enough information nor take too long to train. To address this issue, this paper proposes a deep q-learning network (DQN) with a graph neural network (GNN) as its model. The GNN transforms the input data into a graph which better represents the football players' locations so that it extracts more information of the interactions between different players. With two GNNs to approximate its local and target value functions, this DQN allows players to learn from their experience by using value functions to see the prospective value of each intended action. The proposed model demonstrated the power of GNN in the football game by outperforming other DRL models with significantly fewer steps." }, "2204.13992.pdf": { "id": "2204.13992v1", "title": "A Validation Procedure for the Estimation of Reachable Regions in Football", "year": "2022", "authors": [ "M. Renkin", "J. Bischofberger", "E. Schikuta", "A. Baca" ], "categories": "cs.OH", "pdf_url": "https://arxiv.org/pdf/2204.13992v1", "abstract": "Modelling the trajectorial motion of humans along the ground is a foundational task in the quantitative analysis of sports like association football. Most existing models of football player motion have not been validated yet with respect to actual data. One of the reasons for this lack is that performing such a validation is not straightforward, because models of player motion are usually phrased in a way that emphasises possibly reachable positions rather than expected positions. Since positional data of football players typically contains outliers, this data may misrepresent the range of actually reachable positions. This paper proposes a validation routine for trajectorial motion models that measures and optimises the ability of a motion model to accurately predict all possibly reachable positions by favoring the smallest predicted area of reachable positions that encompasses all observed reached positions up to a manually defined threshold. We demonstrate validation and optimisation on four different motion models, assuming (a) motion with constant speed, (b) motion with constant acceleration, (c) motion with constant acceleration with a speed limit, and (d) motion along two segments with constant speed. Our results show that assuming motion with constant speed or constant acceleration without a limit on the achievable speed is particularly inappropriate for an accurate distinction between reachable and unreachable locations. Motion along two segments of constant speed provides by far the highest accuracy among the tested models and serves as an efficient and accurate approximation of real-world player motion." }, "2205.00254.pdf": { "id": "2205.00254v1", "title": "PGD: A Large-scale Professional Go Dataset for Data-driven Analytics", "year": "2022", "authors": [ "Yifan Gao" ], "categories": "cs.AI, cs.LG", "pdf_url": "https://arxiv.org/pdf/2205.00254v1", "abstract": "Lee Sedol is on a winning streak--does this legend rise again after the competition with AlphaGo? Ke Jie is invincible in the world championship--can he still win the title this time? Go is one of the most popular board games in East Asia, with a stable professional sports system that has lasted for decades in China, Japan, and Korea. There are mature data-driven analysis technologies for many sports, such as soccer, basketball, and esports. However, developing such technology for Go remains nontrivial and challenging due to the lack of datasets, meta-information, and in-game statistics. This paper creates the Professional Go Dataset (PGD), containing 98,043 games played by 2,148 professional players from 1950 to 2021. After manual cleaning and labeling, we provide detailed meta-information for each player, game, and tournament. Moreover, the dataset includes analysis results for each move in the match evaluated by advanced AlphaZero-based AI. To establish a benchmark for PGD, we further analyze the data and extract meaningful in-game features based on prior knowledge related to Go that can indicate the game status. With the help of complete meta-information and constructed in-game features, our results prediction system achieves an accuracy of 75.30%, much higher than several state-of-the-art approaches (64%-65%). As far as we know, PGD is the first dataset for data-driven analytics in Go and even in board games. Beyond this promising result, we provide more examples of tasks that benefit from our dataset. The ultimate goal of this paper is to bridge this ancient game and the modern data science community. It will advance research on Go-related analytics to enhance the fan experience, help players improve their ability, and facilitate other promising aspects. The dataset will be made publicly available." }, "2205.00694.pdf": { "id": "2205.00694v1", "title": "A Multi-stage deep architecture for summary generation of soccer videos", "year": "2022", "authors": [ "Melissa Sanabria", "Fr\u00e9d\u00e9ric Precioso", "Pierre-Alexandre Mattei", "Thomas Menguy" ], "categories": "cs.CV, cs.AI, cs.LG, cs.MM", "pdf_url": "https://arxiv.org/pdf/2205.00694v1", "abstract": "Video content is present in an ever-increasing number of fields, both scientific and commercial. Sports, particularly soccer, is one of the industries that has invested the most in the field of video analytics, due to the massive popularity of the game and the emergence of new markets. Previous state-of-the-art methods on soccer matches video summarization rely on handcrafted heuristics to generate summaries which are poorly generalizable, but these works have yet proven that multiple modalities help detect the best actions of the game. On the other hand, machine learning models with higher generalization potential have entered the field of summarization of general-purpose videos, offering several deep learning approaches. However, most of them exploit content specificities that are not appropriate for sport whole-match videos. Although video content has been for many years the main source for automatizing knowledge extraction in soccer, the data that records all the events happening on the field has become lately very important in sports analytics, since this event data provides richer context information and requires less processing. We propose a method to generate the summary of a soccer match exploiting both the audio and the event metadata. The results show that our method can detect the actions of the match, identify which of these actions should belong to the summary and then propose multiple candidate summaries which are similar enough but with relevant variability to provide different options to the final editor. Furthermore, we show the generalization capability of our work since it can transfer knowledge between datasets from different broadcasting companies, different competitions, acquired in different conditions, and corresponding to summaries of different lengths" }, "2205.02561.pdf": { "id": "2205.02561v3", "title": "LDSA: Learning Dynamic Subtask Assignment in Cooperative Multi-Agent Reinforcement Learning", "year": "2022", "authors": [ "Mingyu Yang", "Jian Zhao", "Xunhan Hu", "Wengang Zhou", "Jiangcheng Zhu", "Houqiang Li" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2205.02561v3", "abstract": "Cooperative multi-agent reinforcement learning (MARL) has made prominent progress in recent years. For training efficiency and scalability, most of the MARL algorithms make all agents share the same policy or value network. However, in many complex multi-agent tasks, different agents are expected to possess specific abilities to handle different subtasks. In those scenarios, sharing parameters indiscriminately may lead to similar behavior across all agents, which will limit the exploration efficiency and degrade the final performance. To balance the training complexity and the diversity of agent behavior, we propose a novel framework to learn dynamic subtask assignment (LDSA) in cooperative MARL. Specifically, we first introduce a subtask encoder to construct a vector representation for each subtask according to its identity. To reasonably assign agents to different subtasks, we propose an ability-based subtask selection strategy, which can dynamically group agents with similar abilities into the same subtask. In this way, agents dealing with the same subtask share their learning of specific abilities and different subtasks correspond to different specific abilities. We further introduce two regularizers to increase the representation difference between subtasks and stabilize the training by discouraging agents from frequently changing subtasks, respectively. Empirical results show that LDSA learns reasonable and effective subtask assignment for better collaboration and significantly improves the learning performance on the challenging StarCraft II micromanagement benchmark and Google Research Football." }, "2205.02731.pdf": { "id": "2205.02731v2", "title": "Characterizing player's playing styles based on Player Vectors for each playing position in the Chinese Football Super League", "year": "2022", "authors": [ "Yuesen Li", "Shouxin Zong", "Yanfei Shen", "Zhiqiang Pu", "Miguel-\u00c1ngel G\u00f3mez", "Yixiong Cui" ], "categories": "cs.LG, stat.AP", "pdf_url": "https://arxiv.org/pdf/2205.02731v2", "abstract": "Characterizing playing style is important for football clubs on scouting, monitoring and match preparation. Previous studies considered a player's style as a combination of technical performances, failing to consider the spatial information. Therefore, this study aimed to characterize the playing styles of each playing position in the Chinese Football Super League (CSL) matches, integrating a recently adopted Player Vectors framework. Data of 960 matches from 2016-2019 CSL were used. Match ratings, and ten types of match events with the corresponding coordinates for all the lineup players whose on-pitch time exceeded 45 minutes were extracted. Players were first clustered into 8 positions. A player vector was constructed for each player in each match based on the Player Vectors using Nonnegative Matrix Factorization (NMF). Another NMF process was run on the player vectors to extract different types of playing styles. The resulting player vectors discovered 18 different playing styles in the CSL. Six performance indicators of each style were investigated to observe their contributions. In general, the playing styles of forwards and midfielders are in line with football performance evolution trends, while the styles of defenders should be reconsidered. Multifunctional playing styles were also found in high rated CSL players." }, "2205.04511.pdf": { "id": "2205.04511v3", "title": "Validation of a motion model for soccer players' sprint by means of tracking data", "year": "2022", "authors": [ "Takuma Narizuka", "Kenta Takizawa", "Yoshihiro Yamazaki" ], "categories": "physics.soc-ph, physics.class-ph", "pdf_url": "https://arxiv.org/pdf/2205.04511v3", "abstract": "In soccer game analysis, the widespread availability of play-by-play and tracking data has made it possible to test mathematical models that have been discussed mainly theoretically. One of the essential models in soccer game analysis is a motion model that predicts the arrival point of a player in $ t $ s. Although many space evaluation and pass prediction methods rely on motion models, the validity of each has not been fully clarified. This study focuses on the motion model proposed by Fujimura and Sugihara (Fujimura-Sugihara model) under sprint conditions based on the equation of motion. A previous study indicated that the Fujimura-Sugihara model is ineffective for soccer games because it generates a circular arrival region. This study aims to examine the validity of the Fujimura-Sugihara model using soccer tracking data. Specifically, we quantitatively compare the arrival regions of players between the model and real data. We show that the boundary of the player's arrival region is circular rather than elliptical, which is consistent with the model. We also show that the initial speed dependence of the arrival region satisfies the solution of the model. Furthermore, we propose a method for estimating valid kinetic parameters in the model directly from tracking data and discuss the limitations of the model for soccer games based on the estimated parameters." }, "2205.05359.pdf": { "id": "2205.05359v3", "title": "Exploring Local Explanations of Nonlinear Models Using Animated Linear Projections", "year": "2022", "authors": [ "Nicholas Spyrison", "Dianne Cook", "Przemyslaw Biecek" ], "categories": "stat.ML, cs.AI, cs.LG", "pdf_url": "https://arxiv.org/pdf/2205.05359v3", "abstract": "The increased predictive power of machine learning models comes at the cost of increased complexity and loss of interpretability, particularly in comparison to parametric statistical models. This trade-off has led to the emergence of eXplainable AI (XAI) which provides methods, such as local explanations (LEs) and local variable attributions (LVAs), to shed light on how a model use predictors to arrive at a prediction. These provide a point estimate of the linear variable importance in the vicinity of a single observation. However, LVAs tend not to effectively handle association between predictors. To understand how the interaction between predictors affects the variable importance estimate, we can convert LVAs into linear projections and use the radial tour. This is also useful for learning how a model has made a mistake, or the effect of outliers, or the clustering of observations. The approach is illustrated with examples from categorical (penguin species, chocolate types) and quantitative (soccer/football salaries, house prices) response models. The methods are implemented in the R package cheem, available on CRAN." }, "2205.10450.pdf": { "id": "2205.10450v2", "title": "Temporally Precise Action Spotting in Soccer Videos Using Dense Detection Anchors", "year": "2022", "authors": [ "Jo\u00e3o V. B. Soares", "Avijit Shah", "Topojoy Biswas" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2205.10450v2", "abstract": "We present a model for temporally precise action spotting in videos, which uses a dense set of detection anchors, predicting a detection confidence and corresponding fine-grained temporal displacement for each anchor. We experiment with two trunk architectures, both of which are able to incorporate large temporal contexts while preserving the smaller-scale features required for precise localization: a one-dimensional version of a u-net, and a Transformer encoder (TE). We also suggest best practices for training models of this kind, by applying Sharpness-Aware Minimization (SAM) and mixup data augmentation. We achieve a new state-of-the-art on SoccerNet-v2, the largest soccer video dataset of its kind, with marked improvements in temporal localization. Additionally, our ablations show: the importance of predicting the temporal displacements; the trade-offs between the u-net and TE trunks; and the benefits of training with SAM and mixup." }, "2205.10953.pdf": { "id": "2205.10953v1", "title": "CYRUS Soccer Simulation 2D Team Description Paper 2022", "year": "2022", "authors": [ "Nader Zare", "Arad Firouzkouhi", "Omid Amini", "Mahtab Sarvmaili", "Aref Sayareh", "Saba Ramezani Rad" ], "categories": "cs.AI, cs.LG, cs.RO", "pdf_url": "https://arxiv.org/pdf/2205.10953v1", "abstract": "Soccer Simulation 2D League is one of the major leagues of RoboCup competitions. In a Soccer Simulation 2D (SS2D) game, two teams of 11 players and one coach compete against each other. The players are only allowed to communicate with the server that is called Soccer Simulation Server. This paper introduces the previous and current research of the CYRUS soccer simulation team, the champion of RoboCup 2021. We will present our idea about improving Unmarking Decisioning and Positioning by using Pass Prediction Deep Neural Network. Based on our experimental results, this idea proven to be effective on increasing the winning rate of Cyrus against opponents." }, "2206.01038.pdf": { "id": "2206.01038v1", "title": "A Survey on Video Action Recognition in Sports: Datasets, Methods and Applications", "year": "2022", "authors": [ "Fei Wu", "Qingzhong Wang", "Jian Bian", "Haoyi Xiong", "Ning Ding", "Feixiang Lu" ], "categories": "cs.CV, cs.AI, cs.MM", "pdf_url": "https://arxiv.org/pdf/2206.01038v1", "abstract": "To understand human behaviors, action recognition based on videos is a common approach. Compared with image-based action recognition, videos provide much more information. Reducing the ambiguity of actions and in the last decade, many works focused on datasets, novel models and learning approaches have improved video action recognition to a higher level. However, there are challenges and unsolved problems, in particular in sports analytics where data collection and labeling are more sophisticated, requiring sport professionals to annotate data. In addition, the actions could be extremely fast and it becomes difficult to recognize them. Moreover, in team sports like football and basketball, one action could involve multiple players, and to correctly recognize them, we need to analyse all players, which is relatively complicated. In this paper, we present a survey on video action recognition for sports analytics. We introduce more than ten types of sports, including team sports, such as football, basketball, volleyball, hockey and individual sports, such as figure skating, gymnastics, table tennis, tennis, diving and badminton. Then we compare numerous existing frameworks for sports analysis to present status quo of video action recognition in both team sports and individual sports. Finally, we discuss the challenges and unsolved problems in this area and to facilitate sports analytics, we develop a toolbox using PaddlePaddle, which supports football, basketball, table tennis and figure skating action recognition." }, "2206.01899.pdf": { "id": "2206.01899v3", "title": "Evaluation of creating scoring opportunities for teammates in soccer via trajectory prediction", "year": "2022", "authors": [ "Masakiyo Teranishi", "Kazushi Tsutsui", "Kazuya Takeda", "Keisuke Fujii" ], "categories": "cs.AI, cs.LG, cs.MA", "pdf_url": "https://arxiv.org/pdf/2206.01899v3", "abstract": "Evaluating the individual movements for teammates in soccer players is crucial for assessing teamwork, scouting, and fan engagement. It has been said that players in a 90-min game do not have the ball for about 87 minutes on average. However, it has remained difficult to evaluate an attacking player without receiving the ball, and to reveal how movement contributes to the creation of scoring opportunities for teammates. In this paper, we evaluate players who create off-ball scoring opportunities by comparing actual movements with the reference movements generated via trajectory prediction. First, we predict the trajectories of players using a graph variational recurrent neural network that can accurately model the relationship between players and predict the long-term trajectory. Next, based on the difference in the modified off-ball evaluation index between the actual and the predicted trajectory as a reference, we evaluate how the actual movement contributes to scoring opportunity compared to the predicted movement. For verification, we examined the relationship with the annual salary, the goals, and the rating in the game by experts for all games of a team in a professional soccer league in a year. The results show that the annual salary and the proposed indicator correlated significantly, which could not be explained by the existing indicators and goals. Our results suggest the effectiveness of the proposed method as an indicator for a player without the ball to create a scoring chance for teammates." }, "2206.02310.pdf": { "id": "2206.02310v1", "title": "CYRUS Soccer Simulation 2D Team Description Paper 2021", "year": "2022", "authors": [ "Nader Zare", "Aref Sayareh", "Mahtab Sarvmaili", "Omid Amini", "Amilcar Soares", "Stan Matwin" ], "categories": "cs.RO", "pdf_url": "https://arxiv.org/pdf/2206.02310v1", "abstract": "In this report, we briefly present the technical procedure and simulation steps for the 2D soccer simulation of team Cyrus. We emphasize on this document on how the prediction of teammates' behavior is performed. In our proposed method, the agent receives the noisy inputs from the server, and predicts the ball holder full state behavior. Taking advantage of this approach for choosing the optimal view angle shows 11.30% improvement on the expected win rate." }, "2206.07212.pdf": { "id": "2206.07212v2", "title": "Explainable expected goal models for performance analysis in football analytics", "year": "2022", "authors": [ "Mustafa Cavus", "Przemys\u0142aw Biecek" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2206.07212v2", "abstract": "The expected goal provides a more representative measure of the team and player performance which also suit the low-scoring nature of football instead of score in modern football. The score of a match involves randomness and often may not represent the performance of the teams and players, therefore it has been popular to use the alternative statistics in recent years such as shots on target, ball possessions, and drills. To measure the probability of a shot being a goal by the expected goal, several features are used to train an expected goal model which is based on the event and tracking football data. The selection of these features, the size and date of the data, and the model which are used as the parameters that may affect the performance of the model. Using black-box machine learning models for increasing the predictive performance of the model decreases its interpretability that causes the loss of information that can be gathered from the model. This paper proposes an accurate expected goal model trained consisting of 315,430 shots from seven seasons between 2014-15 and 2020-21 of the top-five European football leagues. Moreover, this model is explained by using explainable artificial intelligence tool to obtain an explainable expected goal model for evaluating a team or player performance. To the best of our knowledge, this is the first paper that demonstrates a practical application of an explainable artificial intelligence tool aggregated profiles to explain a group of observations on an accurate expected goal model for monitoring the team and player performance. Moreover, these methods can be generalized to other sports branches." }, "2206.10926.pdf": { "id": "2206.10926v2", "title": "SoccerCPD: Formation and Role Change-Point Detection in Soccer Matches Using Spatiotemporal Tracking Data", "year": "2022", "authors": [ "Hyunsung Kim", "Bit Kim", "Dongwook Chung", "Jinsung Yoon", "Sang-Ki Ko" ], "categories": "stat.AP, stat.ML", "pdf_url": "https://arxiv.org/pdf/2206.10926v2", "abstract": "In fluid team sports such as soccer and basketball, analyzing team formation is one of the most intuitive ways to understand tactics from domain participants' point of view. However, existing approaches either assume that team formation is consistent throughout a match or assign formations frame-by-frame, which disagree with real situations. To tackle this issue, we propose a change-point detection framework named SoccerCPD that distinguishes tactically intended formation and role changes from temporary changes in soccer matches. We first assign roles to players frame-by-frame and perform two-step change-point detections: (1) formation change-point detection based on the sequence of role-adjacency matrices and (2) role change-point detection based on the sequence of role permutations. The evaluation of SoccerCPD using the ground truth annotated by domain experts shows that our method accurately detects the points of tactical changes and estimates the formation and role assignment per segment. Lastly, we introduce practical use-cases that domain participants can easily interpret and utilize." }, "2206.11335.pdf": { "id": "2206.11335v1", "title": "Video Analytics in Elite Soccer: A Distributed Computing Perspective", "year": "2022", "authors": [ "Debesh Jha", "Ashish Rauniyar", "H\u00e5vard D. Johansen", "Dag Johansen", "Michael A. Riegler", "P\u00e5l Halvorsen" ], "categories": "cs.MM", "pdf_url": "https://arxiv.org/pdf/2206.11335v1", "abstract": "Ubiquitous sensors and Internet of Things (IoT) technologies have revolutionized the sports industry, providing new methodologies for planning, effective coordination of training, and match analysis post game. New methods, including machine learning, image and video processing, have been developed for performance evaluation, allowing the analyst to track the performance of a player in real-time. Following FIFA's 2015 approval of electronics performance and tracking system during games, performance data of a single player or the entire team is allowed to be collected using GPS-based wearables. Data from practice sessions outside the sporting arena is being collected in greater numbers than ever before. Realizing the significance of data in professional soccer, this paper presents video analytics, examines recent state-of-the-art literature in elite soccer, and summarizes existing real-time video analytics algorithms. We also discuss real-time crowdsourcing of the obtained data, tactical and technical performance, distributed computing and its importance in video analytics and propose a future research perspective." }, "2206.13246.pdf": { "id": "2206.13246v1", "title": "Prediction of Football Player Value using Bayesian Ensemble Approach", "year": "2022", "authors": [ "Hansoo Lee", "Bayu Adhi Tama", "Meeyoung Cha" ], "categories": "cs.LG, cs.AI, cs.IR", "pdf_url": "https://arxiv.org/pdf/2206.13246v1", "abstract": "The transfer fees of sports players have become astronomical. This is because bringing players of great future value to the club is essential for their survival. We present a case study on the key factors affecting the world's top soccer players' transfer fees based on the FIFA data analysis. To predict each player's market value, we propose an improved LightGBM model by optimizing its hyperparameter using a Tree-structured Parzen Estimator (TPE) algorithm. We identify prominent features by the SHapley Additive exPlanations (SHAP) algorithm. The proposed method has been compared against the baseline regression models (e.g., linear regression, lasso, elastic net, kernel ridge regression) and gradient boosting model without hyperparameter optimization. The optimized LightGBM model showed an excellent accuracy of approximately 3.8, 1.4, and 1.8 times on average compared to the regression baseline models, GBDT, and LightGBM model in terms of RMSE. Our model offers interpretability in deciding what attributes football clubs should consider in recruiting players in the future." }, "2207.05691.pdf": { "id": "2207.05691v2", "title": "The MuSe 2022 Multimodal Sentiment Analysis Challenge: Humor, Emotional Reactions, and Stress", "year": "2022", "authors": [ "Lukas Christ", "Shahin Amiriparian", "Alice Baird", "Panagiotis Tzirakis", "Alexander Kathan", "Niklas M\u00fcller" ], "categories": "cs.LG, cs.AI, cs.CL, cs.MM, eess.AS", "pdf_url": "https://arxiv.org/pdf/2207.05691v2", "abstract": "The Multimodal Sentiment Analysis Challenge (MuSe) 2022 is dedicated to multimodal sentiment and emotion recognition. For this year's challenge, we feature three datasets: (i) the Passau Spontaneous Football Coach Humor (Passau-SFCH) dataset that contains audio-visual recordings of German football coaches, labelled for the presence of humour; (ii) the Hume-Reaction dataset in which reactions of individuals to emotional stimuli have been annotated with respect to seven emotional expression intensities, and (iii) the Ulm-Trier Social Stress Test (Ulm-TSST) dataset comprising of audio-visual data labelled with continuous emotion values (arousal and valence) of people in stressful dispositions. Using the introduced datasets, MuSe 2022 2022 addresses three contemporary affective computing problems: in the Humor Detection Sub-Challenge (MuSe-Humor), spontaneous humour has to be recognised; in the Emotional Reactions Sub-Challenge (MuSe-Reaction), seven fine-grained `in-the-wild' emotions have to be predicted; and in the Emotional Stress Sub-Challenge (MuSe-Stress), a continuous prediction of stressed emotion values is featured. The challenge is designed to attract different research communities, encouraging a fusion of their disciplines. Mainly, MuSe 2022 targets the communities of audio-visual emotion recognition, health informatics, and symbolic sentiment analysis. This baseline paper describes the datasets as well as the feature sets extracted from them. A recurrent neural network with LSTM cells is used to set competitive baseline results on the test partitions for each sub-challenge. We report an Area Under the Curve (AUC) of .8480 for MuSe-Humor; .2801 mean (from 7-classes) Pearson's Correlations Coefficient for MuSe-Reaction, as well as .4931 Concordance Correlation Coefficient (CCC) and .4761 for valence and arousal in MuSe-Stress, respectively." }, "2207.08017.pdf": { "id": "2207.08017v2", "title": "Is Soccer a lie or simply a complex system?", "year": "2022", "authors": [ "Nelson Fernandez", "Ricardo Bernal" ], "categories": "cs.SI, physics.soc-ph", "pdf_url": "https://arxiv.org/pdf/2207.08017v2", "abstract": "Understanding soccer as a complex system we base on nature and the collective behavior of many organisms that \"do calculations,\" seeking to generate solutions in a bioinspired way. When soccer mysteries appear, complex systems science emerges as a means to provide explanations. However, given the variety of interpretations that complexity and its associated properties can have and the understanding of what a complex system is, it is convenient to provide some elements to understand how unpredictability in soccer gives way to hundreds of counterintuitive results and how the science of complexity could contribute to the understanding of many phenomena in this sport. In this context, the manuscript's objective is to synthetically address some of the most important aspects of applied complexity to soccer to bring science and sport closer together" }, "2207.11361.pdf": { "id": "2207.11361v1", "title": "Machine Learning Modeling to Evaluate the Value of Football Players", "year": "2022", "authors": [ "Chenyao Li", "Stylianos Kampakis", "Philip Treleaven" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2207.11361v1", "abstract": "In most sports, especially football, most coaches and analysts search for key performance indicators using notational analysis. This method utilizes a statistical summary of events based on video footage and numerical records of goal scores. Unfortunately, this approach is now obsolete owing to the continuous evolutionary increase in technology that simplifies the analysis of more complex process variables through machine learning (ML). Machine learning, a form of artificial intelligence (AI), uses algorithms to detect meaningful patterns and define a structure based on positional data. This research investigates a new method to evaluate the value of current football players, based on establishing the machine learning models to investigate the relations among the various features of players, the salary of players, and the market value of players. The data of the football players used for this project is from several football websites. The data on the salary of football players will be the proxy for evaluating the value of players, and other features will be used to establish and train the ML model for predicting the suitable salary for the players. The motivation is to explore what are the relations between different features of football players and their salaries - how each feature affects their salaries, or which are the most important features to affect the salary? Although many standards can reflect the value of football players, the salary of the players is one of the most intuitive and crucial indexes, so this study will use the salary of players as the proxy to evaluate their value. Moreover, many features of players can affect the valuation of the football players, but the value of players is mainly decided by three types of factors: basic characteristics, performance on the court, and achievements at the club." }, "2207.12710.pdf": { "id": "2207.12710v2", "title": "Active Learning of Ordinal Embeddings: A User Study on Football Data", "year": "2022", "authors": [ "Christoffer Loeffler", "Kion Fallah", "Stefano Fenu", "Dario Zanca", "Bjoern Eskofier", "Christopher John Rozell" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2207.12710v2", "abstract": "Humans innately measure distance between instances in an unlabeled dataset using an unknown similarity function. Distance metrics can only serve as proxy for similarity in information retrieval of similar instances. Learning a good similarity function from human annotations improves the quality of retrievals. This work uses deep metric learning to learn these user-defined similarity functions from few annotations for a large football trajectory dataset. We adapt an entropy-based active learning method with recent work from triplet mining to collect easy-to-answer but still informative annotations from human participants and use them to train a deep convolutional network that generalizes to unseen samples. Our user study shows that our approach improves the quality of the information retrieval compared to a previous deep metric learning approach that relies on a Siamese network. Specifically, we shed light on the strengths and weaknesses of passive sampling heuristics and active learners alike by analyzing the participants' response efficacy. To this end, we collect accuracy, algorithmic time complexity, the participants' fatigue and time-to-response, qualitative self-assessment and statements, as well as the effects of mixed-expertise annotators and their consistency on model performance and transfer-learning." }, "2208.01160.pdf": { "id": "2208.01160v1", "title": "Hierarchical Reinforcement Learning for Precise Soccer Shooting Skills using a Quadrupedal Robot", "year": "2022", "authors": [ "Yandong Ji", "Zhongyu Li", "Yinan Sun", "Xue Bin Peng", "Sergey Levine", "Glen Berseth" ], "categories": "cs.RO, cs.AI, eess.SY", "pdf_url": "https://arxiv.org/pdf/2208.01160v1", "abstract": "We address the problem of enabling quadrupedal robots to perform precise shooting skills in the real world using reinforcement learning. Developing algorithms to enable a legged robot to shoot a soccer ball to a given target is a challenging problem that combines robot motion control and planning into one task. To solve this problem, we need to consider the dynamics limitation and motion stability during the control of a dynamic legged robot. Moreover, we need to consider motion planning to shoot the hard-to-model deformable ball rolling on the ground with uncertain friction to a desired location. In this paper, we propose a hierarchical framework that leverages deep reinforcement learning to train (a) a robust motion control policy that can track arbitrary motions and (b) a planning policy to decide the desired kicking motion to shoot a soccer ball to a target. We deploy the proposed framework on an A1 quadrupedal robot and enable it to accurately shoot the ball to random targets in the real world." }, "2209.03882.pdf": { "id": "2209.03882v1", "title": "Valuing Players Over Time", "year": "2022", "authors": [ "Tiago Mendes-Neves", "Lu\u00eds Meireles", "Jo\u00e3o Mendes-Moreira" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2209.03882v1", "abstract": "In soccer (or association football), players quickly go from heroes to zeroes, or vice-versa. Performance is not a static measure but a somewhat volatile one. Analyzing performance as a time series rather than a stationary point in time is crucial to making better decisions. This paper introduces and explores I-VAEP and O-VAEP models to evaluate actions and rate players' intention and execution. Then, we analyze these ratings over time and propose use cases to fundament our option of treating player ratings as a continuous problem. As a result, we present who were the best players and how their performance evolved, define volatility metrics to measure a player's consistency, and build a player development curve to assist decision-making." }, "2209.09491.pdf": { "id": "2209.09491v2", "title": "Deep Q-Network for AI Soccer", "year": "2022", "authors": [ "Curie Kim", "Yewon Hwang", "Jong-Hwan Kim" ], "categories": "cs.AI", "pdf_url": "https://arxiv.org/pdf/2209.09491v2", "abstract": "Reinforcement learning has shown an outstanding performance in the applications of games, particularly in Atari games as well as Go. Based on these successful examples, we attempt to apply one of the well-known reinforcement learning algorithms, Deep Q-Network, to the AI Soccer game. AI Soccer is a 5:5 robot soccer game where each participant develops an algorithm that controls five robots in a team to defeat the opponent participant. Deep Q-Network is designed to implement our original rewards, the state space, and the action space to train each agent so that it can take proper actions in different situations during the game. Our algorithm was able to successfully train the agents, and its performance was preliminarily proven through the mini-competition against 10 teams wishing to take part in the AI Soccer international competition. The competition was organized by the AI World Cup committee, in conjunction with the WCG 2019 Xi'an AI Masters. With our algorithm, we got the achievement of advancing to the round of 16 in this international competition with 130 teams from 39 countries." }, "2209.10055.pdf": { "id": "2209.10055v1", "title": "Lamarckian Platform: Pushing the Boundaries of Evolutionary Reinforcement Learning towards Asynchronous Commercial Games", "year": "2022", "authors": [ "Hui Bai", "Ruimin Shen", "Yue Lin", "Botian Xu", "Ran Cheng" ], "categories": "cs.LG, cs.AI, cs.NE", "pdf_url": "https://arxiv.org/pdf/2209.10055v1", "abstract": "Despite the emerging progress of integrating evolutionary computation into reinforcement learning, the absence of a high-performance platform endowing composability and massive parallelism causes non-trivial difficulties for research and applications related to asynchronous commercial games. Here we introduce Lamarckian - an open-source platform featuring support for evolutionary reinforcement learning scalable to distributed computing resources. To improve the training speed and data efficiency, Lamarckian adopts optimized communication methods and an asynchronous evolutionary reinforcement learning workflow. To meet the demand for an asynchronous interface by commercial games and various methods, Lamarckian tailors an asynchronous Markov Decision Process interface and designs an object-oriented software architecture with decoupled modules. In comparison with the state-of-the-art RLlib, we empirically demonstrate the unique advantages of Lamarckian on benchmark tests with up to 6000 CPU cores: i) both the sampling efficiency and training speed are doubled when running PPO on Google football game; ii) the training speed is 13 times faster when running PBT+PPO on Pong game. Moreover, we also present two use cases: i) how Lamarckian is applied to generating behavior-diverse game AI; ii) how Lamarckian is applied to game balancing tests for an asynchronous commercial game." }, "2209.11949.pdf": { "id": "2209.11949v1", "title": "Hybrid Multimodal Fusion for Humor Detection", "year": "2022", "authors": [ "Haojie Xu", "Weifeng Liu", "Jingwei Liu", "Mingzheng Li", "Yu Feng", "Yasi Peng" ], "categories": "cs.LG, cs.AI, cs.CL, cs.MM", "pdf_url": "https://arxiv.org/pdf/2209.11949v1", "abstract": "In this paper, we present our solution to the MuSe-Humor sub-challenge of the Multimodal Emotional Challenge (MuSe) 2022. The goal of the MuSe-Humor sub-challenge is to detect humor and calculate AUC from audiovisual recordings of German football Bundesliga press conferences. It is annotated for humor displayed by the coaches. For this sub-challenge, we first build a discriminant model using the transformer module and BiLSTM module, and then propose a hybrid fusion strategy to use the prediction results of each modality to improve the performance of the model. Our experiments demonstrate the effectiveness of our proposed model and hybrid fusion strategy on multimodal fusion, and the AUC of our proposed model on the test set is 0.8972." }, "2209.13154.pdf": { "id": "2209.13154v1", "title": "Observation Centric and Central Distance Recovery on Sports Player Tracking", "year": "2022", "authors": [ "Hsiang-Wei Huang", "Cheng-Yen Yang", "Jenq-Neng Hwang", "Pyong-Kun Kim", "Kwangju Kim", "Kyoungoh Lee" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2209.13154v1", "abstract": "Multi-Object Tracking over humans has improved rapidly with the development of object detection and re-identification. However, multi-actor tracking over humans with similar appearance and nonlinear movement can still be very challenging even for the state-of-the-art tracking algorithm. Current motion-based tracking algorithms often use Kalman Filter to predict the motion of an object, however, its linear movement assumption can cause failure in tracking when the target is not moving linearly. And for multi-players tracking over the sports field, because the players in the same team are usually wearing the same color of jersey, making re-identification even harder both in the short term and long term in the tracking process. In this work, we proposed a motionbased tracking algorithm and three post-processing pipelines for three sports including basketball, football, and volleyball, we successfully handle the tracking of the non-linear movement of players on the sports fields. Experiments result on the testing set of ECCV DeeperAction Challenge SportsMOT Dataset demonstrate the effectiveness of our method, which achieves a HOTA of 73.968, ranking 3rd place on the 2022 Sportsmot workshop final leaderboard." }, "2209.13254.pdf": { "id": "2209.13254v1", "title": "Identifying and Extracting Football Features from Real-World Media Sources using Only Synthetic Training Data", "year": "2022", "authors": [ "Jose Cerqueira Fernandes", "Benjamin Kenwright" ], "categories": "cs.AI, cs.GR", "pdf_url": "https://arxiv.org/pdf/2209.13254v1", "abstract": "Real-world images used for training machine learning algorithms are often unstructured and inconsistent. The process of analysing and tagging these images can be costly and error prone (also availability, gaps and legal conundrums). However, as we demonstrate in this article, the potential to generate accurate graphical images that are indistinguishable from real-world sources has a multitude of benefits in machine learning paradigms. One such example of this is football data from broadcast services (television and other streaming media sources). The football games are usually recorded from multiple sources (cameras and phones) and resolutions, not to mention, occlusion of visual details and other artefacts (like blurring, weathering and lighting conditions) which make it difficult to accurately identify features. We demonstrate an approach which is able to overcome these limitations using generated tagged and structured images. The generated images are able to simulate a variety views and conditions (including noise and blurring) which may only occur sporadically in real-world data and make it difficult for machine learning algorithm to 'cope' with these unforeseen problems in real-data. This approach enables us to rapidly train and prepare a robust solution that accurately extracts features (e.g., spacial locations, markers on the pitch, player positions, ball location and camera FOV) from real-world football match sources for analytical purposes." }, "2209.13846.pdf": { "id": "2209.13846v2", "title": "VREN: Volleyball Rally Dataset with Expression Notation Language", "year": "2022", "authors": [ "Haotian Xia", "Rhys Tracy", "Yun Zhao", "Erwan Fraisse", "Yuan-Fang Wang", "Linda Petzold" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2209.13846v2", "abstract": "This research is intended to accomplish two goals: The first goal is to curate a large and information rich dataset that contains crucial and succinct summaries on the players' actions and positions and the back-and-forth travel patterns of the volleyball in professional and NCAA Div-I indoor volleyball games. While several prior studies have aimed to create similar datasets for other sports (e.g. badminton and soccer), creating such a dataset for indoor volleyball is not yet realized. The second goal is to introduce a volleyball descriptive language to fully describe the rally processes in the games and apply the language to our dataset. Based on the curated dataset and our descriptive sports language, we introduce three tasks for automated volleyball action and tactic analysis using our dataset: (1) Volleyball Rally Prediction, aimed at predicting the outcome of a rally and helping players and coaches improve decision-making in practice, (2) Setting Type and Hitting Type Prediction, to help coaches and players prepare more effectively for the game, and (3) Volleyball Tactics and Attacking Zone Statistics, to provide advanced volleyball statistics and help coaches understand the game and opponent's tactics better. We conducted case studies to show how experimental results can provide insights to the volleyball analysis community. Furthermore, experimental evaluation based on real-world data establishes a baseline for future studies and applications of our dataset and language. This study bridges the gap between the indoor volleyball field and computer science. The dataset is available at: https://github.com/haotianxia/VREN." }, "2210.06327.pdf": { "id": "2210.06327v3", "title": "Betting the system: Using lineups to predict football scores", "year": "2022", "authors": [ "George Peters", "Diogo Pacheco" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2210.06327v3", "abstract": "This paper aims to reduce randomness in football by analysing the role of lineups in final scores using machine learning prediction models we have developed. Football clubs invest millions of dollars on lineups and knowing how individual statistics translate to better outcomes can optimise investments. Moreover, sports betting is growing exponentially and being able to predict the future is profitable and desirable. We use machine learning models and historical player data from English Premier League (2020-2022) to predict scores and to understand how individual performance can improve the outcome of a match. We compared different prediction techniques to maximise the possibility of finding useful models. We created heuristic and machine learning models predicting football scores to compare different techniques. We used different sets of features and shown goalkeepers stats are more important than attackers stats to predict goals scored. We applied a broad evaluation process to assess the efficacy of the models in real world applications. We managed to predict correctly all relegated teams after forecast 100 consecutive matches. We show that Support Vector Regression outperformed other techniques predicting final scores and that lineups do not improve predictions. Finally, our model was profitable (42% return) when emulating a betting system using real world odds data." }, "2210.08872.pdf": { "id": "2210.08872v2", "title": "PTDE: Personalized Training with Distilled Execution for Multi-Agent Reinforcement Learning", "year": "2022", "authors": [ "Yiqun Chen", "Hangyu Mao", "Jiaxin Mao", "Shiguang Wu", "Tianle Zhang", "Bin Zhang" ], "categories": "cs.AI, cs.LG, cs.MA", "pdf_url": "https://arxiv.org/pdf/2210.08872v2", "abstract": "Centralized Training with Decentralized Execution (CTDE) has emerged as a widely adopted paradigm in multi-agent reinforcement learning, emphasizing the utilization of global information for learning an enhanced joint $Q$-function or centralized critic. In contrast, our investigation delves into harnessing global information to directly enhance individual $Q$-functions or individual actors. Notably, we discover that applying identical global information universally across all agents proves insufficient for optimal performance. Consequently, we advocate for the customization of global information tailored to each agent, creating agent-personalized global information to bolster overall performance. Furthermore, we introduce a novel paradigm named Personalized Training with Distilled Execution (PTDE), wherein agent-personalized global information is distilled into the agent's local information. This distilled information is then utilized during decentralized execution, resulting in minimal performance degradation. PTDE can be seamlessly integrated with state-of-the-art algorithms, leading to notable performance enhancements across diverse benchmarks, including the SMAC benchmark, Google Research Football (GRF) benchmark, and Learning to Rank (LTR) task." }, "2210.11802.pdf": { "id": "2210.11802v1", "title": "Postural balance asymmetry and subsequent noncontact lower extremity musculoskeletal injuries among Tunisian soccer players with groin pain: A prospective case control study", "year": "2022", "authors": [ "Fatma Chaari", "S\u00e9bastien Boyas", "Sonia Sahli", "Thouraya Fendri", "Mohammed A Harrabi", "Haithem Rebai" ], "categories": "physics.med-ph", "pdf_url": "https://arxiv.org/pdf/2210.11802v1", "abstract": "Background: Recent studies reported postural balance disorders in patients and soccer players with groin pain (GP) compared to controls. Since postural balance asymmetry identified after an initial injury contributes for subsequent injuries, identification of this asymmetry in soccer players with GP may highlight the risk of sustaining subsequent noncontact lower extremity musculoskeletal injuries in these players. Therefore, the aims of this study were to (i) examine static and dynamic unipedal postural balance asymmetry in soccer players with GP compared to healthy ones, and (ii) quantify the risk of subsequent noncontact lower extremity injuries in these players. Research question: Do soccer players with GP exhibit higher static and dynamic unipedal postural balance asymmetry, and higher risk of sustaining subsequent injuries compared to controls. Methods: In this prospective case control study, 27 soccer players with non-time loss GP (GP group: GPG), and 27 healthy ones (control group: CG) were enrolled. Static and dynamic unipedal postural balance asymmetry were evaluated with a force platform using symmetry index (SI), and Y-balance test (Y-BT), respectively. Additionally, subsequent noncontact lower extremity musculoskeletal injuries were tracked for 10 months. Results: The GPG revealed higher (p < 0.01) SI in eyes closed condition, higher (p < 0.001) side-to-side asymmetry in anterior, posteromedial and posterolateral reach distances and in composite Y-BT score compared to CG. They showed lower (p < 0.001) composite score for injured limb and higher (p < 0.001) side-to-side asymmetry in posteromedial reach distance compared to the cutoff values of 89.6 % and 4 cm, respectively. Moreover, GPG exhibited higher odds (OR= 7.48; 95 % CI = 2.15, 26.00; p < 0.01) of sustaining subsequent injuries compared to CG. Significance: The Y-BT should be instituted into existing pre-participation physical examinations to screen for soccer players with non-time loss GP at an elevated risk of sustaining subsequent injuries. This could help coaches and clinicians make valid return to play decisions." }, "2210.12262.pdf": { "id": "2210.12262v1", "title": "Group Distributionally Robust Reinforcement Learning with Hierarchical Latent Variables", "year": "2022", "authors": [ "Mengdi Xu", "Peide Huang", "Yaru Niu", "Visak Kumar", "Jielin Qiu", "Chao Fang" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2210.12262v1", "abstract": "One key challenge for multi-task Reinforcement learning (RL) in practice is the absence of task indicators. Robust RL has been applied to deal with task ambiguity, but may result in over-conservative policies. To balance the worst-case (robustness) and average performance, we propose Group Distributionally Robust Markov Decision Process (GDR-MDP), a flexible hierarchical MDP formulation that encodes task groups via a latent mixture model. GDR-MDP identifies the optimal policy that maximizes the expected return under the worst-possible qualified belief over task groups within an ambiguity set. We rigorously show that GDR-MDP's hierarchical structure improves distributional robustness by adding regularization to the worst possible outcomes. We then develop deep RL algorithms for GDR-MDP for both value-based and policy-based RL methods. Extensive experiments on Box2D control tasks, MuJoCo benchmarks, and Google football platforms show that our algorithms outperform classic robust training algorithms across diverse environments in terms of robustness under belief uncertainties. Demos are available on our project page (\\url{https://sites.google.com/view/gdr-rl/home})." }, "2210.17368.pdf": { "id": "2210.17368v1", "title": "Teacher-student curriculum learning for reinforcement learning", "year": "2022", "authors": [ "Yanick Schraner" ], "categories": "cs.LG, cs.AI", "pdf_url": "https://arxiv.org/pdf/2210.17368v1", "abstract": "Reinforcement learning (rl) is a popular paradigm for sequential decision making problems. The past decade's advances in rl have led to breakthroughs in many challenging domains such as video games, board games, robotics, and chip design. The sample inefficiency of deep reinforcement learning methods is a significant obstacle when applying rl to real-world problems. Transfer learning has been applied to reinforcement learning such that the knowledge gained in one task can be applied when training in a new task. Curriculum learning is concerned with sequencing tasks or data samples such that knowledge can be transferred between those tasks to learn a target task that would otherwise be too difficult to solve. Designing a curriculum that improves sample efficiency is a complex problem. In this thesis, we propose a teacher-student curriculum learning setting where we simultaneously train a teacher that selects tasks for the student while the student learns how to solve the selected task. Our method is independent of human domain knowledge and manual curriculum design. We evaluated our methods on two reinforcement learning benchmarks: grid world and the challenging Google Football environment. With our method, we can improve the sample efficiency and generality of the student compared to tabula-rasa reinforcement learning." }, "2211.00374.pdf": { "id": "2211.00374v1", "title": "What drives a goalkeepers' decisions?", "year": "2022", "authors": [ "Samer Fatayri", "Kirill Serykh", "Egor Gumin" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2211.00374v1", "abstract": "In soccer games, the goalkeeper's performance is an important factor to the success of the whole team. Despite the goalkeeper's importance, little attention has been paid to their performance in events and tracking data. Here, we developed a model to predict which movements would be most effective for shot-stopping and compare it to the real-life behavior of goalkeepers. This model evaluates the performance of goalkeepers based on their position and dive radius. We found that contrary to the movements that were considered most effective by our model, real-life goalkeepers' movements were more diverse. We further used our model to develop a tool to analyse goalkeepers' behavior in real-life soccer games. In addition, a simulator function allows team analysts or couches to identify situations that allow further improvement of the reaction of the goalkeeper." }, "2211.08585.pdf": { "id": "2211.08585v1", "title": "Cyrus2D base: Source Code Base for RoboCup 2D Soccer Simulation League", "year": "2022", "authors": [ "Nader Zare", "Omid Amini", "Aref Sayareh", "Mahtab Sarvmaili", "Arad Firouzkouhi", "Saba Ramezani Rad" ], "categories": "cs.RO", "pdf_url": "https://arxiv.org/pdf/2211.08585v1", "abstract": "Soccer Simulation 2D League is one of the major leagues of RoboCup competitions. In a Soccer Simulation 2D (SS2D) game, two teams of 11 players and one coach compete against each other. Several base codes have been released for the RoboCup soccer simulation 2D (RCSS2D) community that have promoted the application of multi-agent and AI algorithms in this field. In this paper, we introduce \"Cyrus2D Base\", which is derived from the base code of the RCSS2D 2021 champion. We merged Gliders2D base V2.6 with the newest version of the Helios base. We applied several features of Cyrus2021 to improve the performance and capabilities of this base alongside a Data Extractor to facilitate the implementation of machine learning in the field. We have tested this base code in different teams and scenarios, and the obtained results demonstrate significant improvements in the defensive and offensive strategy of the team." }, "2211.15734.pdf": { "id": "2211.15734v1", "title": "Predicting Football Match Outcomes with eXplainable Machine Learning and the Kelly Index", "year": "2022", "authors": [ "Yiming Ren", "Teo Susnjak" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2211.15734v1", "abstract": "In this work, a machine learning approach is developed for predicting the outcomes of football matches. The novelty of this research lies in the utilisation of the Kelly Index to first classify matches into categories where each one denotes the different levels of predictive difficulty. Classification models using a wide suite of algorithms were developed for each category of matches in order to determine the efficacy of the approach. In conjunction to this, a set of previously unexplored features were engineering including Elo-based variables. The dataset originated from the Premier League match data covering the 2019-2021 seasons. The findings indicate that the process of decomposing the predictive problem into sub-tasks was effective and produced competitive results with prior works, while the ensemble-based methods were the most effective. The paper also devised an investment strategy in order to evaluate its effectiveness by benchmarking against bookmaker odds. An approach was developed that minimises risk by combining the Kelly Index with the predefined confidence thresholds of the predictive models. The experiments found that the proposed strategy can return a profit when following a conservative approach that focuses primarily on easy-to-predict matches where the predictive models display a high confidence level." }, "2212.00021.pdf": { "id": "2212.00021v1", "title": "Location analysis of players in UEFA EURO 2020 and 2022 using generalized valuation of defense by estimating probabilities", "year": "2022", "authors": [ "Rikuhei Umemoto", "Kazushi Tsutsui", "Keisuke Fujii" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2212.00021v1", "abstract": "Analyzing defenses in team sports is generally challenging because of the limited event data. Researchers have previously proposed methods to evaluate football team defense by predicting the events of ball gain and being attacked using locations of all players and the ball. However, they did not consider the importance of the events, assumed the perfect observation of all 22 players, and did not fully investigated the influence of the diversity (e.g., nationality and sex). Here, we propose a generalized valuation method of defensive teams by score-scaling the predicted probabilities of the events. Using the open-source location data of all players in broadcast video frames in football games of men's Euro 2020 and women's Euro 2022, we investigated the effect of the number of players on the prediction and validated our approach by analyzing the games. Results show that for the predictions of being attacked, scoring, and conceding, all players' information was not necessary, while that of ball gain required information on three to four offensive and defensive players. With game analyses we explained the excellence in defense of finalist teams in Euro 2020. Our approach might be applicable to location data from broadcast video frames in football games." }, "2212.00389.pdf": { "id": "2212.00389v1", "title": "Kick-motion Training with DQN in AI Soccer Environment", "year": "2022", "authors": [ "Bumgeun Park", "Jihui Lee", "Taeyoung Kim", "Dongsoo Har" ], "categories": "cs.RO, cs.AI", "pdf_url": "https://arxiv.org/pdf/2212.00389v1", "abstract": "This paper presents a technique to train a robot to perform kick-motion in AI soccer by using reinforcement learning (RL). In RL, an agent interacts with an environment and learns to choose an action in a state at each step. When training RL algorithms, a problem called the curse of dimensionality (COD) can occur if the dimension of the state is high and the number of training data is low. The COD often causes degraded performance of RL models. In the situation of the robot kicking the ball, as the ball approaches the robot, the robot chooses the action based on the information obtained from the soccer field. In order not to suffer COD, the training data, which are experiences in the case of RL, should be collected evenly from all areas of the soccer field over (theoretically infinite) time. In this paper, we attempt to use the relative coordinate system (RCS) as the state for training kick-motion of robot agent, instead of using the absolute coordinate system (ACS). Using the RCS eliminates the necessity for the agent to know all the (state) information of entire soccer field and reduces the dimension of the state that the agent needs to know to perform kick-motion, and consequently alleviates COD. The training based on the RCS is performed with the widely used Deep Q-network (DQN) and tested in the AI Soccer environment implemented with Webots simulation software." }, "2212.07397.pdf": { "id": "2212.07397v1", "title": "Hierarchical Strategies for Cooperative Multi-Agent Reinforcement Learning", "year": "2022", "authors": [ "Majd Ibrahim", "Ammar Fayad" ], "categories": "cs.LG, cs.MA", "pdf_url": "https://arxiv.org/pdf/2212.07397v1", "abstract": "Adequate strategizing of agents behaviors is essential to solving cooperative MARL problems. One intuitively beneficial yet uncommon method in this domain is predicting agents future behaviors and planning accordingly. Leveraging this point, we propose a two-level hierarchical architecture that combines a novel information-theoretic objective with a trajectory prediction model to learn a strategy. To this end, we introduce a latent policy that learns two types of latent strategies: individual $z_A$, and relational $z_R$ using a modified Graph Attention Network module to extract interaction features. We encourage each agent to behave according to the strategy by conditioning its local $Q$ functions on $z_A$, and we further equip agents with a shared $Q$ function that conditions on $z_R$. Additionally, we introduce two regularizers to allow predicted trajectories to be accurate and rewarding. Empirical results on Google Research Football (GRF) and StarCraft (SC) II micromanagement tasks show that our method establishes a new state of the art being, to the best of our knowledge, the first MARL algorithm to solve all super hard SC II scenarios as well as the GRF full game with a win rate higher than $95\\%$, thus outperforming all existing methods. Videos and brief overview of the methods and results are available at: https://sites.google.com/view/hier-strats-marl/home." }, "2212.09832.pdf": { "id": "2212.09832v1", "title": "Denoising instrumented mouthguard measurements of head impact kinematics with a convolutional neural network", "year": "2022", "authors": [ "Xianghao Zhan", "Yuzhe Liu", "Nicholas J. Cecchi", "Ashlyn A. Callan", "Enora Le Flao", "Olivier Gevaert" ], "categories": "cs.LG, eess.SP, q-bio.QM, q-bio.TO", "pdf_url": "https://arxiv.org/pdf/2212.09832v1", "abstract": "Wearable sensors for measuring head kinematics can be noisy due to imperfect interfaces with the body. Mouthguards are used to measure head kinematics during impacts in traumatic brain injury (TBI) studies, but deviations from reference kinematics can still occur due to potential looseness. In this study, deep learning is used to compensate for the imperfect interface and improve measurement accuracy. A set of one-dimensional convolutional neural network (1D-CNN) models was developed to denoise mouthguard kinematics measurements along three spatial axes of linear acceleration and angular velocity. The denoised kinematics had significantly reduced errors compared to reference kinematics, and reduced errors in brain injury criteria and tissue strain and strain rate calculated via finite element modeling. The 1D-CNN models were also tested on an on-field dataset of college football impacts and a post-mortem human subject dataset, with similar denoising effects observed. The models can be used to improve detection of head impacts and TBI risk evaluation, and potentially extended to other sensors measuring kinematics." }, "2301.13052.pdf": { "id": "2301.13052v2", "title": "A Machine Learning Approach for Player and Position Adjusted Expected Goals in Football (Soccer)", "year": "2023", "authors": [ "James H. Hewitt", "Oktay Karaku\u015f" ], "categories": "cs.LG, stat.AP", "pdf_url": "https://arxiv.org/pdf/2301.13052v2", "abstract": "Football is a very result-driven industry, with goals being rarer than in most sports, so having further parameters to judge the performance of teams and individuals is key. Expected Goals (xG) allow further insight than just a scoreline. To tackle the need for further analysis in football, this paper uses machine learning applications that are developed and applied to Football Event data. From the concept, a Binary Classification problem is created whereby a probabilistic valuation is outputted using Logistic Regression and Gradient Boosting based approaches. The model successfully predicts xGs probability values for football players based on 15,575 shots. The proposed solution utilises StatsBomb as the data provider and an industry benchmark to tune the models in the right direction. The proposed ML solution for xG is further used to tackle the age-old cliche of: 'the ball has fallen to the wrong guy there'. The development of the model is used to adjust and gain more realistic values of expected goals than the general models show. To achieve this, this paper tackles Positional Adjusted xG, splitting the training data into Forward, Midfield, and Defence with the aim of providing insight into player qualities based on their positional sub-group. Positional Adjusted xG successfully predicts and proves that more attacking players are better at accumulating xG. The highest value belonged to Forwards followed by Midfielders and Defenders. Finally, this study has further developments into Player Adjusted xG with the aim of proving that Messi is statistically at a higher efficiency level than the average footballer. This is achieved by using Messi subset samples to quantify his qualities in comparison to the average xG models finding that Messi xG performs 347 xG higher than the general model outcome." }, "2302.00123.pdf": { "id": "2302.00123v1", "title": "Design and Implementation of A Soccer Ball Detection System with Multiple Cameras", "year": "2023", "authors": [ "Lei Li", "Tianfang Zhang", "Zhongfeng Kang", "Wenhan Zhang" ], "categories": "cs.CV, cs.AI", "pdf_url": "https://arxiv.org/pdf/2302.00123v1", "abstract": "The detection of small and medium-sized objects in three dimensions has always been a frontier exploration problem. This technology has a very wide application in sports analysis, games, virtual reality, human animation and other fields. The traditional three-dimensional small target detection technology has the disadvantages of high cost, low precision and inconvenience, so it is difficult to apply in practice. With the development of machine learning and deep learning, the technology of computer vision algorithms is becoming more mature. Creating an immersive media experience is considered to be a very important research work in sports. The main work is to explore and solve the problem of football detection under the multiple cameras, aiming at the research and implementation of the live broadcast system of football matches. Using multi cameras detects a target ball and determines its position in three dimension with the occlusion, motion, low illumination of the target object. This paper designed and implemented football detection system under multiple cameras for the detection and capture of targets in real-time matches. The main work mainly consists of three parts, football detector, single camera detection, and multi-cameras detection. The system used bundle adjustment to obtain the three-dimensional position of the target, and the GPU to accelerates data pre-processing and achieve accurate real-time capture of the target. By testing the system, it shows that the system can accurately detect and capture the moving targets in 3D. In addition, the solution in this paper is reusable for large-scale competitions, like basketball and soccer. The system framework can be well transplanted into other similar engineering project systems. It has been put into the market." }, "2302.03426.pdf": { "id": "2302.03426v1", "title": "Impact of velocity and impact angle on football shot accuracy during fundamental trainings", "year": "2023", "authors": [ "Rahman Sahinler", "Omer Burak Goktas", "Berkay Mumcu", "Damla Sen", "Feyza Kocaturk", "Huseyin Uvet" ], "categories": "eess.SP", "pdf_url": "https://arxiv.org/pdf/2302.03426v1", "abstract": "The purpose of this research is to create a machine learning-based smart coaching approach for football that can replace manual analysis with real-time feedback for trainers. In-depth analysis of football player data by humans is time-consuming, error-prone, and requires a lot of effort. This exploratory study demonstrates the feasibility of using a machine learning algorithm to enhance the effectiveness of player monitoring and training. The suggested approach uses machine learning to generate analytical insights and enable long-term monitoring of player performance. In the future, machine learning could use this technique to offer constructive criticism of football players. The system incorporates a homemade ball-throwing mechanism capable of launching the ball in a variety of directions and at varying velocities. The ball kicker is equipped with a gyroscope and accelerometer sensors for measuring velocity and acceleration. The gathered data is filtered initially, and then the data that has been processed is fed into the machine-learning algorithm. The algorithm will be trained on player performance data and will be able to provide real-time feedback to coaches on player performance and potential areas for improvement. Additionally, the system will be able to track player progress over time and provide coaches with a comprehensive view of player development. The ultimate goal is to improve player performance and reduce the workload for coaches by automating the analysis process." }, "2302.06569.pdf": { "id": "2302.06569v1", "title": "Inferring Player Location in Sports Matches: Multi-Agent Spatial Imputation from Limited Observations", "year": "2023", "authors": [ "Gregory Everett", "Ryan J. Beal", "Tim Matthews", "Joseph Early", "Timothy J. Norman", "Sarvapali D. Ramchurn" ], "categories": "cs.LG, cs.MA", "pdf_url": "https://arxiv.org/pdf/2302.06569v1", "abstract": "Understanding agent behaviour in Multi-Agent Systems (MAS) is an important problem in domains such as autonomous driving, disaster response, and sports analytics. Existing MAS problems typically use uniform timesteps with observations for all agents. In this work, we analyse the problem of agent location imputation, specifically posed in environments with non-uniform timesteps and limited agent observability (~95% missing values). Our approach uses Long Short-Term Memory and Graph Neural Network components to learn temporal and inter-agent patterns to predict the location of all agents at every timestep. We apply this to the domain of football (soccer) by imputing the location of all players in a game from sparse event data (e.g., shots and passes). Our model estimates player locations to within ~6.9m; a ~62% reduction in error from the best performing baseline. This approach facilitates downstream analysis tasks such as player physical metrics, player coverage, and team pitch control. Existing solutions to these tasks often require optical tracking data, which is expensive to obtain and only available to elite clubs. By imputing player locations from easy to obtain event data, we increase the accessibility of downstream tasks." }, "2302.07515.pdf": { "id": "2302.07515v2", "title": "TiZero: Mastering Multi-Agent Football with Curriculum Learning and Self-Play", "year": "2023", "authors": [ "Fanqi Lin", "Shiyu Huang", "Tim Pearce", "Wenze Chen", "Wei-Wei Tu" ], "categories": "cs.AI, cs.LG, cs.MA", "pdf_url": "https://arxiv.org/pdf/2302.07515v2", "abstract": "Multi-agent football poses an unsolved challenge in AI research. Existing work has focused on tackling simplified scenarios of the game, or else leveraging expert demonstrations. In this paper, we develop a multi-agent system to play the full 11 vs. 11 game mode, without demonstrations. This game mode contains aspects that present major challenges to modern reinforcement learning algorithms; multi-agent coordination, long-term planning, and non-transitivity. To address these challenges, we present TiZero; a self-evolving, multi-agent system that learns from scratch. TiZero introduces several innovations, including adaptive curriculum learning, a novel self-play strategy, and an objective that optimizes the policies of multiple agents jointly. Experimentally, it outperforms previous systems by a large margin on the Google Research Football environment, increasing win rates by over 30%. To demonstrate the generality of TiZero's innovations, they are assessed on several environments beyond football; Overcooked, Multi-agent Particle-Environment, Tic-Tac-Toe and Connect-Four." }, "2302.09276.pdf": { "id": "2302.09276v1", "title": "Transformer-Based Neural Marked Spatio Temporal Point Process Model for Football Match Events Analysis", "year": "2023", "authors": [ "Calvin C. K. Yeung", "Tony Sit", "Keisuke Fujii" ], "categories": "cs.AI, cs.LG", "pdf_url": "https://arxiv.org/pdf/2302.09276v1", "abstract": "With recently available football match event data that record the details of football matches, analysts and researchers have a great opportunity to develop new performance metrics, gain insight, and evaluate key performance. However, most sports sequential events modeling methods and performance metrics approaches could be incomprehensive in dealing with such large-scale spatiotemporal data (in particular, temporal process), thereby necessitating a more comprehensive spatiotemporal model and a holistic performance metric. To this end, we proposed the Transformer-Based Neural Marked Spatio Temporal Point Process (NMSTPP) model for football event data based on the neural temporal point processes (NTPP) framework. In the experiments, our model outperformed the prediction performance of the baseline models. Furthermore, we proposed the holistic possession utilization score (HPUS) metric for a more comprehensive football possession analysis. For verification, we examined the relationship with football teams' final ranking, average goal score, and average xG over a season. It was observed that the average HPUS showed significant correlations regardless of not using goal and details of shot information. Furthermore, we show HPUS examples in analyzing possessions, matches, and between matches." }, "2302.10760.pdf": { "id": "2302.10760v1", "title": "Potential Penetrative Pass (P3)", "year": "2023", "authors": [ "Hadi Sotudeh" ], "categories": "cs.LG, cs.AI", "pdf_url": "https://arxiv.org/pdf/2302.10760v1", "abstract": "To score goals in football, a team needs to move forward on the pitch and there are various ways to do so. Depending on the game plan & philosophy; some teams prefer to play long balls from either wings or defense. Others, prefer to penetrate in depth with passes and outplay the opponent players. To objectively & in an automated way evaluate how teams play penetrative passes compared to the number of times they had the potential to do so, the \"Potential Penetrative Pass (P3)\" concept is presented here." }, "2302.14058.pdf": { "id": "2302.14058v1", "title": "Identification of pattern mining algorithm for rugby league players positional groups separation based on movement patterns", "year": "2023", "authors": [ "Victor Elijah Adeyemo", "Anna Palczewska", "Ben Jones", "Dan Weaving" ], "categories": "cs.LG, cs.AI", "pdf_url": "https://arxiv.org/pdf/2302.14058v1", "abstract": "The application of pattern mining algorithms to extract movement patterns from sports big data can improve training specificity by facilitating a more granular evaluation of movement. As there are various pattern mining algorithms, this study aimed to validate which algorithm discovers the best set of movement patterns for player movement profiling in professional rugby league and the similarity in extracted movement patterns between the algorithms. Three pattern mining algorithms (l-length Closed Contiguous [LCCspm], Longest Common Subsequence [LCS] and AprioriClose) were used to profile elite rugby football league hookers (n = 22 players) and wingers (n = 28 players) match-games movements across 319 matches. Machine learning classification algorithms were used to identify which algorithm gives the best set of movement patterns to separate playing positions with Jaccard similarity score identifying the extent of similarity between algorithms' movement patterns. LCCspm and LCS movement patterns shared a 0.19 Jaccard similarity score. AprioriClose movement patterns shared no significant similarity with LCCspm and LCS patterns. The closed contiguous movement patterns profiled by LCCspm best-separated players into playing positions. Multi-layered Perceptron algorithm achieved the highest accuracy of 91.02% and precision, recall and F1 scores of 0.91 respectively. Therefore, we recommend the extraction of closed contiguous (consecutive) over non-consecutive movement patterns for separating groups of players." }, "2303.01318.pdf": { "id": "2303.01318v2", "title": "A Continuous-Time Stochastic Process for High-Resolution Network Data in Sports", "year": "2023", "authors": [ "Nicholas Grieshop", "Yong Feng", "Guanyu Hu", "Michael Schweinberger" ], "categories": "stat.AP", "pdf_url": "https://arxiv.org/pdf/2303.01318v2", "abstract": "Technological advances have paved the way for collecting high-resolution network data in basketball, football, and other team-based sports. Such data consist of interactions among players of competing teams indexed by space and time. High-resolution network data are vital to understanding and predicting the performance of teams, because the performance of a team is more than the sum of the strengths of its individual players: Whether a collection of players forms a strong team depends on the strength of the individual players as well as the interactions among the players. We introduce a continuous-time stochastic process as a model of interactions among players of competing teams indexed by space and time, discuss basic properties of the continuous-time stochastic process, and learn the stochastic process from high-resolution network data by pursuing a Bayesian approach. We present simulation results along with an application to Juventus Turin, Inter Milan, and other football clubs in the premier Italian soccer league." }, "2303.12401.pdf": { "id": "2303.12401v2", "title": "Real-time forecasting within soccer matches through a Bayesian lens", "year": "2023", "authors": [ "Chinmay Divekar", "Soudeep Deb", "Rishideep Roy" ], "categories": "stat.AP, stat.ME", "pdf_url": "https://arxiv.org/pdf/2303.12401v2", "abstract": "This paper employs a Bayesian methodology to predict the results of soccer matches in real-time. Using sequential data of various events throughout the match, we utilize a multinomial probit regression in a novel framework to estimate the time-varying impact of covariates and to forecast the outcome. English Premier League data from eight seasons are used to evaluate the efficacy of our method. Different evaluation metrics establish that the proposed model outperforms potential competitors inspired by existing statistical or machine learning algorithms. Additionally, we apply robustness checks to demonstrate the model's accuracy across various scenarios." }, "2303.13323.pdf": { "id": "2303.13323v1", "title": "Deep Generative Multi-Agent Imitation Model as a Computational Benchmark for Evaluating Human Performance in Complex Interactive Tasks: A Case Study in Football", "year": "2023", "authors": [ "Chaoyi Gu", "Varuna De Silva" ], "categories": "stat.ML, cs.LG, cs.MA", "pdf_url": "https://arxiv.org/pdf/2303.13323v1", "abstract": "Evaluating the performance of human is a common need across many applications, such as in engineering and sports. When evaluating human performance in completing complex and interactive tasks, the most common way is to use a metric having been proved efficient for that context, or to use subjective measurement techniques. However, this can be an error prone and unreliable process since static metrics cannot capture all the complex contexts associated with such tasks and biases exist in subjective measurement. The objective of our research is to create data-driven AI agents as computational benchmarks to evaluate human performance in solving difficult tasks involving multiple humans and contextual factors. We demonstrate this within the context of football performance analysis. We train a generative model based on Conditional Variational Recurrent Neural Network (VRNN) Model on a large player and ball tracking dataset. The trained model is used to imitate the interactions between two teams and predict the performance from each team. Then the trained Conditional VRNN Model is used as a benchmark to evaluate team performance. The experimental results on Premier League football dataset demonstrates the usefulness of our method to existing state-of-the-art static metric used in football analytics." }, "2303.15305.pdf": { "id": "2303.15305v1", "title": "Why winning a soccer game is worth 5 points", "year": "2023", "authors": [ "Leszek Szczecinski" ], "categories": "physics.soc-ph", "pdf_url": "https://arxiv.org/pdf/2303.15305v1", "abstract": "Ranking is used in sport leagues to determine a champion and/or to decide on promotion/relegation of teams. Arguably, the best known ranking method relies on scores obtained by cumulating the points associated with the wins and the draws of all teams, which are then ranked by sorting the score obtained. There are two main problems with this ranking method. First, the meaning of the ranking is undefined, and, second, it depends on the relative value of the wins that is arbitrarily set. We remedy these issues by introducing a probabilistic model of the game results and by showing an interpretation of the ranking that is consistent with the model. We also propose a methodology to estimate the parameter of the model which allows us to objectively determine the value of the win. In particular, using data from the association football (soccer), we show that the value of the win is close to five (5) points." }, "2303.15471.pdf": { "id": "2303.15471v3", "title": "Embedding Contextual Information through Reward Shaping in Multi-Agent Learning: A Case Study from Google Football", "year": "2023", "authors": [ "Chaoyi Gu", "Varuna De Silva", "Corentin Artaud", "Rafael Pina" ], "categories": "cs.LG, cs.AI", "pdf_url": "https://arxiv.org/pdf/2303.15471v3", "abstract": "Artificial Intelligence has been used to help human complete difficult tasks in complicated environments by providing optimized strategies for decision-making or replacing the manual labour. In environments including multiple agents, such as football, the most common methods to train agents are Imitation Learning and Multi-Agent Reinforcement Learning (MARL). However, the agents trained by Imitation Learning cannot outperform the expert demonstrator, which makes humans hardly get new insights from the learnt policy. Besides, MARL is prone to the credit assignment problem. In environments with sparse reward signal, this method can be inefficient. The objective of our research is to create a novel reward shaping method by embedding contextual information in reward function to solve the aforementioned challenges. We demonstrate this in the Google Research Football (GRF) environment. We quantify the contextual information extracted from game state observation and use this quantification together with original sparse reward to create the shaped reward. The experiment results in the GRF environment prove that our reward shaping method is a useful addition to state-of-the-art MARL algorithms for training agents in environments with sparse reward signal." }, "2303.16648.pdf": { "id": "2303.16648v1", "title": "Beating the average: how to generate profit by exploiting the inefficiencies of soccer betting", "year": "2023", "authors": [ "Ralph St\u00f6mmer" ], "categories": "physics.soc-ph, math.PR, physics.pop-ph", "pdf_url": "https://arxiv.org/pdf/2303.16648v1", "abstract": "In economy, markets are denoted as efficient when it is impossible to systematically generate profits which outperform the average. In the past years, the concept has been tested in other domains such as the growing sports betting market. Surprisingly, despite its large size and its level of maturity, sports betting shows traits of inefficiency. The anomalies indicate the existence of strategies which shift betting from a game of chance towards a game of skill. This article shows an example for an inefficiency detected in the German soccer betting TOTO 13er Wette, which is operated by state-run lottery agencies. Gamblers have to guess the outcome (win, draw, loss) of 13 soccer matches listed on a lottery tip. Applying stochastic methods, a recipe is presented to determine hit rates for single match outcomes. More important, the recipe provides the number of lottery tips required to achieve a specific number of strikes (number of correct match forecasts per lottery tip) for any given level of safety. An approximation is derived to cope with large numbers in hypergeometric distributions, valid under certain constraints. Overall, the strategy does lead to returns exceeding the aggregated lottery fees, resulting in moderate, but consistent profits. It is briefly discussed if lessions learned from soccer betting can be transferred back to financial markets, because gamblers and retail investors face similar challenges and opportunities." }, "2303.17707.pdf": { "id": "2303.17707v4", "title": "Why is plausibility surprisingly problematic as an XAI criterion?", "year": "2023", "authors": [ "Weina Jin", "Xiaoxiao Li", "Ghassan Hamarneh" ], "categories": "cs.AI, cs.HC", "pdf_url": "https://arxiv.org/pdf/2303.17707v4", "abstract": "Explainable artificial intelligence (XAI) is motivated by the problem of making AI predictions understandable, transparent, and responsible, as AI becomes increasingly impactful in society and high-stakes domains. The evaluation and optimization criteria of XAI are gatekeepers for XAI algorithms to achieve their expected goals and should withstand rigorous inspection. To improve the scientific rigor of XAI, we conduct a critical examination of a common XAI criterion: plausibility. Plausibility assesses how convincing the AI explanation is to humans, and is usually quantified by metrics of feature localization or feature correlation. Our examination shows that plausibility is invalid to measure explainability, and human explanations are not the ground truth for XAI, because doing so ignores the necessary assumptions underpinning an explanation. Our examination further reveals the consequences of using plausibility as an XAI criterion, including increasing misleading explanations that manipulate users, deteriorating users' trust in the AI system, undermining human autonomy, being unable to achieve complementary human-AI task performance, and abandoning other possible approaches of enhancing understandability. Due to the invalidity of measurements and the unethical issues, this position paper argues that the community should stop using plausibility as a criterion for the evaluation and optimization of XAI algorithms. We also delineate new research approaches to improve XAI in trustworthiness, understandability, and utility to users, including complementary human-AI task performance." }, "2304.04220.pdf": { "id": "2304.04220v1", "title": "Towards Active Learning for Action Spotting in Association Football Videos", "year": "2023", "authors": [ "Silvio Giancola", "Anthony Cioppa", "Julia Georgieva", "Johsan Billingham", "Andreas Serner", "Kerry Peek" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2304.04220v1", "abstract": "Association football is a complex and dynamic sport, with numerous actions occurring simultaneously in each game. Analyzing football videos is challenging and requires identifying subtle and diverse spatio-temporal patterns. Despite recent advances in computer vision, current algorithms still face significant challenges when learning from limited annotated data, lowering their performance in detecting these patterns. In this paper, we propose an active learning framework that selects the most informative video samples to be annotated next, thus drastically reducing the annotation effort and accelerating the training of action spotting models to reach the highest accuracy at a faster pace. Our approach leverages the notion of uncertainty sampling to select the most challenging video clips to train on next, hastening the learning process of the algorithm. We demonstrate that our proposed active learning framework effectively reduces the required training data for accurate action spotting in football videos. We achieve similar performances for action spotting with NetVLAD++ on SoccerNet-v2, using only one-third of the dataset, indicating significant capabilities for reducing annotation time and improving data efficiency. We further validate our approach on two new datasets that focus on temporally localizing actions of headers and passes, proving its effectiveness across different action semantics in football. We believe our active learning framework for action spotting would support further applications of action spotting algorithms and accelerate annotation campaigns in the sports domain." }, "2304.04565.pdf": { "id": "2304.04565v1", "title": "SoccerNet-Caption: Dense Video Captioning for Soccer Broadcasts Commentaries", "year": "2023", "authors": [ "Hassan Mkhallati", "Anthony Cioppa", "Silvio Giancola", "Bernard Ghanem", "Marc Van Droogenbroeck" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2304.04565v1", "abstract": "Soccer is more than just a game - it is a passion that transcends borders and unites people worldwide. From the roar of the crowds to the excitement of the commentators, every moment of a soccer match is a thrill. Yet, with so many games happening simultaneously, fans cannot watch them all live. Notifications for main actions can help, but lack the engagement of live commentary, leaving fans feeling disconnected. To fulfill this need, we propose in this paper a novel task of dense video captioning focusing on the generation of textual commentaries anchored with single timestamps. To support this task, we additionally present a challenging dataset consisting of almost 37k timestamped commentaries across 715.9 hours of soccer broadcast videos. Additionally, we propose a first benchmark and baseline for this task, highlighting the difficulty of temporally anchoring commentaries yet showing the capacity to generate meaningful commentaries. By providing broadcasters with a tool to summarize the content of their video with the same level of engagement as a live game, our method could help satisfy the needs of the numerous fans who follow their team but cannot necessarily watch the live game. We believe our method has the potential to enhance the accessibility and understanding of soccer content for a wider audience, bringing the excitement of the game to more people." }, "2304.04617.pdf": { "id": "2304.04617v1", "title": "VARS: Video Assistant Referee System for Automated Soccer Decision Making from Multiple Views", "year": "2023", "authors": [ "Jan Held", "Anthony Cioppa", "Silvio Giancola", "Abdullah Hamdi", "Bernard Ghanem", "Marc Van Droogenbroeck" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2304.04617v1", "abstract": "The Video Assistant Referee (VAR) has revolutionized association football, enabling referees to review incidents on the pitch, make informed decisions, and ensure fairness. However, due to the lack of referees in many countries and the high cost of the VAR infrastructure, only professional leagues can benefit from it. In this paper, we propose a Video Assistant Referee System (VARS) that can automate soccer decision-making. VARS leverages the latest findings in multi-view video analysis, to provide real-time feedback to the referee, and help them make informed decisions that can impact the outcome of a game. To validate VARS, we introduce SoccerNet-MVFoul, a novel video dataset of soccer fouls from multiple camera views, annotated with extensive foul descriptions by a professional soccer referee, and we benchmark our VARS to automatically recognize the characteristics of these fouls. We believe that VARS has the potential to revolutionize soccer refereeing and take the game to new heights of fairness and accuracy across all levels of professional and amateur federations." }, "2304.13653.pdf": { "id": "2304.13653v2", "title": "Learning Agile Soccer Skills for a Bipedal Robot with Deep Reinforcement Learning", "year": "2023", "authors": [ "Tuomas Haarnoja", "Ben Moran", "Guy Lever", "Sandy H. Huang", "Dhruva Tirumala", "Jan Humplik" ], "categories": "cs.RO, cs.AI, cs.LG", "pdf_url": "https://arxiv.org/pdf/2304.13653v2", "abstract": "We investigate whether Deep Reinforcement Learning (Deep RL) is able to synthesize sophisticated and safe movement skills for a low-cost, miniature humanoid robot that can be composed into complex behavioral strategies in dynamic environments. We used Deep RL to train a humanoid robot with 20 actuated joints to play a simplified one-versus-one (1v1) soccer game. The resulting agent exhibits robust and dynamic movement skills such as rapid fall recovery, walking, turning, kicking and more; and it transitions between them in a smooth, stable, and efficient manner. The agent's locomotion and tactical behavior adapts to specific game contexts in a way that would be impractical to manually design. The agent also developed a basic strategic understanding of the game, and learned, for instance, to anticipate ball movements and to block opponent shots. Our agent was trained in simulation and transferred to real robots zero-shot. We found that a combination of sufficiently high-frequency control, targeted dynamics randomization, and perturbations during training in simulation enabled good-quality transfer. Although the robots are inherently fragile, basic regularization of the behavior during training led the robots to learn safe and effective movements while still performing in a dynamic and agile way -- well beyond what is intuitively expected from the robot. Indeed, in experiments, they walked 181% faster, turned 302% faster, took 63% less time to get up, and kicked a ball 34% faster than a scripted baseline, while efficiently combining the skills to achieve the longer term objectives." }, "2305.09458.pdf": { "id": "2305.09458v1", "title": "An Empirical Study on Google Research Football Multi-agent Scenarios", "year": "2023", "authors": [ "Yan Song", "He Jiang", "Zheng Tian", "Haifeng Zhang", "Yingping Zhang", "Jiangcheng Zhu" ], "categories": "cs.LG, cs.MA", "pdf_url": "https://arxiv.org/pdf/2305.09458v1", "abstract": "Few multi-agent reinforcement learning (MARL) research on Google Research Football (GRF) focus on the 11v11 multi-agent full-game scenario and to the best of our knowledge, no open benchmark on this scenario has been released to the public. In this work, we fill the gap by providing a population-based MARL training pipeline and hyperparameter settings on multi-agent football scenario that outperforms the bot with difficulty 1.0 from scratch within 2 million steps. Our experiments serve as a reference for the expected performance of Independent Proximal Policy Optimization (IPPO), a state-of-the-art multi-agent reinforcement learning algorithm where each agent tries to maximize its own policy independently across various training configurations. Meanwhile, we open-source our training framework Light-MALib which extends the MALib codebase by distributed and asynchronized implementation with additional analytical tools for football games. Finally, we provide guidance for building strong football AI with population-based training and release diverse pretrained policies for benchmarking. The goal is to provide the community with a head start for whoever experiment their works on GRF and a simple-to-use population-based training framework for further improving their agents through self-play. The implementation is available at https://github.com/Shanghai-Digital-Brain-Laboratory/DB-Football." }, "2305.13030.pdf": { "id": "2305.13030v4", "title": "Adaptive action supervision in reinforcement learning from real-world multi-agent demonstrations", "year": "2023", "authors": [ "Keisuke Fujii", "Kazushi Tsutsui", "Atom Scott", "Hiroshi Nakahara", "Naoya Takeishi", "Yoshinobu Kawahara" ], "categories": "cs.AI, cs.LG", "pdf_url": "https://arxiv.org/pdf/2305.13030v4", "abstract": "Modeling of real-world biological multi-agents is a fundamental problem in various scientific and engineering fields. Reinforcement learning (RL) is a powerful framework to generate flexible and diverse behaviors in cyberspace; however, when modeling real-world biological multi-agents, there is a domain gap between behaviors in the source (i.e., real-world data) and the target (i.e., cyberspace for RL), and the source environment parameters are usually unknown. In this paper, we propose a method for adaptive action supervision in RL from real-world demonstrations in multi-agent scenarios. We adopt an approach that combines RL and supervised learning by selecting actions of demonstrations in RL based on the minimum distance of dynamic time warping for utilizing the information of the unknown source dynamics. This approach can be easily applied to many existing neural network architectures and provide us with an RL model balanced between reproducibility as imitation and generalization ability to obtain rewards in cyberspace. In the experiments, using chase-and-escape and football tasks with the different dynamics between the unknown source and target environments, we show that our approach achieved a balance between the reproducibility and the generalization ability compared with the baselines. In particular, we used the tracking data of professional football players as expert demonstrations in football and show successful performances despite the larger gap between behaviors in the source and target environments than the chase-and-escape task." }, "2305.17886.pdf": { "id": "2305.17886v2", "title": "Action valuation of on- and off-ball soccer players based on multi-agent deep reinforcement learning", "year": "2023", "authors": [ "Hiroshi Nakahara", "Kazushi Tsutsui", "Kazuya Takeda", "Keisuke Fujii" ], "categories": "cs.AI, cs.LG", "pdf_url": "https://arxiv.org/pdf/2305.17886v2", "abstract": "Analysis of invasive sports such as soccer is challenging because the game situation changes continuously in time and space, and multiple agents individually recognize the game situation and make decisions. Previous studies using deep reinforcement learning have often considered teams as a single agent and valued the teams and players who hold the ball in each discrete event. Then it was challenging to value the actions of multiple players, including players far from the ball, in a spatiotemporally continuous state space. In this paper, we propose a method of valuing possible actions for on- and off-ball soccer players in a single holistic framework based on multi-agent deep reinforcement learning. We consider a discrete action space in a continuous state space that mimics that of Google research football and leverages supervised learning for actions in reinforcement learning. In the experiment, we analyzed the relationships with conventional indicators, season goals, and game ratings by experts, and showed the effectiveness of the proposed method. Our approach can assess how multiple players move continuously throughout the game, which is difficult to be discretized or labeled but vital for teamwork, scouting, and fan engagement." }, "2305.19283.pdf": { "id": "2305.19283v1", "title": "Observation Denoising in CYRUS Soccer Simulation 2D Team For RoboCup 2023", "year": "2023", "authors": [ "Aref Sayareh", "Nader Zare", "Omid Amini", "Arad Firouzkouhi", "Mahtab Sarvmaili", "Stan Matwin" ], "categories": "cs.AI, cs.RO", "pdf_url": "https://arxiv.org/pdf/2305.19283v1", "abstract": "The RoboCup competitions hold various leagues, and the Soccer Simulation 2D League is a major one among them. Soccer Simulation 2D (SS2D) match involves two teams, including 11 players and a coach, competing against each other. The players can only communicate with the Soccer Simulation Server during the game. This paper presents the latest research of the CYRUS soccer simulation 2D team, the champion of RoboCup 2021. We will explain our denoising idea powered by long short-term memory networks (LSTM) and deep neural networks (DNN). The CYRUS team uses the CYRUS2D base code that was developed based on the Helios and Gliders bases." }, "2306.05255.pdf": { "id": "2306.05255v1", "title": "Toward more accurate and generalizable brain deformation estimators for traumatic brain injury detection with unsupervised domain adaptation", "year": "2023", "authors": [ "Xianghao Zhan", "Jiawei Sun", "Yuzhe Liu", "Nicholas J. Cecchi", "Enora Le Flao", "Olivier Gevaert" ], "categories": "cs.LG, eess.SP, physics.bio-ph, q-bio.QM, stat.AP", "pdf_url": "https://arxiv.org/pdf/2306.05255v1", "abstract": "Machine learning head models (MLHMs) are developed to estimate brain deformation for early detection of traumatic brain injury (TBI). However, the overfitting to simulated impacts and the lack of generalizability caused by distributional shift of different head impact datasets hinders the broad clinical applications of current MLHMs. We propose brain deformation estimators that integrates unsupervised domain adaptation with a deep neural network to predict whole-brain maximum principal strain (MPS) and MPS rate (MPSR). With 12,780 simulated head impacts, we performed unsupervised domain adaptation on on-field head impacts from 302 college football (CF) impacts and 457 mixed martial arts (MMA) impacts using domain regularized component analysis (DRCA) and cycle-GAN-based methods. The new model improved the MPS/MPSR estimation accuracy, with the DRCA method significantly outperforming other domain adaptation methods in prediction accuracy (p<0.001): MPS RMSE: 0.027 (CF) and 0.037 (MMA); MPSR RMSE: 7.159 (CF) and 13.022 (MMA). On another two hold-out test sets with 195 college football impacts and 260 boxing impacts, the DRCA model significantly outperformed the baseline model without domain adaptation in MPS and MPSR estimation accuracy (p<0.001). The DRCA domain adaptation reduces the MPS/MPSR estimation error to be well below TBI thresholds, enabling accurate brain deformation estimation to detect TBI in future clinical applications." }, "2306.08206.pdf": { "id": "2306.08206v1", "title": "Ball Trajectory Inference from Multi-Agent Sports Contexts Using Set Transformer and Hierarchical Bi-LSTM", "year": "2023", "authors": [ "Hyunsung Kim", "Han-Jun Choi", "Chang Jo Kim", "Jinsung Yoon", "Sang-Ki Ko" ], "categories": "cs.MA, cs.AI", "pdf_url": "https://arxiv.org/pdf/2306.08206v1", "abstract": "As artificial intelligence spreads out to numerous fields, the application of AI to sports analytics is also in the spotlight. However, one of the major challenges is the difficulty of automated acquisition of continuous movement data during sports matches. In particular, it is a conundrum to reliably track a tiny ball on a wide soccer pitch with obstacles such as occlusion and imitations. Tackling the problem, this paper proposes an inference framework of ball trajectory from player trajectories as a cost-efficient alternative to ball tracking. We combine Set Transformers to get permutation-invariant and equivariant representations of the multi-agent contexts with a hierarchical architecture that intermediately predicts the player ball possession to support the final trajectory inference. Also, we introduce the reality loss term and postprocessing to secure the estimated trajectories to be physically realistic. The experimental results show that our model provides natural and accurate trajectories as well as admissible player ball possession at the same time. Lastly, we suggest several practical applications of our framework including missing trajectory imputation, semi-automated pass annotation, automated zoom-in for match broadcasting, and calculating possession-wise running performance metrics." }, "2306.16061.pdf": { "id": "2306.16061v2", "title": "MRHER: Model-based Relay Hindsight Experience Replay for Sequential Object Manipulation Tasks with Sparse Rewards", "year": "2023", "authors": [ "Yuming Huang", "Bin Ren", "Ziming Xu", "Lianghong Wu" ], "categories": "cs.RO, cs.AI", "pdf_url": "https://arxiv.org/pdf/2306.16061v2", "abstract": "Sparse rewards pose a significant challenge to achieving high sample efficiency in goal-conditioned reinforcement learning (RL). Specifically, in sequential manipulation tasks, the agent receives failure rewards until it successfully completes the entire manipulation task, which leads to low sample efficiency. To tackle this issue and improve sample efficiency, we propose a novel model-based RL framework called Model-based Relay Hindsight Experience Replay (MRHER). MRHER breaks down a continuous task into subtasks with increasing complexity and utilizes the previous subtask to guide the learning of the subsequent one. Instead of using Hindsight Experience Replay (HER) in every subtask, we design a new robust model-based relabeling method called Foresight relabeling (FR). FR predicts the future trajectory of the hindsight state and relabels the expected goal as a goal achieved on the virtual future trajectory. By incorporating FR, MRHER effectively captures more information from historical experiences, leading to improved sample efficiency, particularly in object-manipulation environments. Experimental results demonstrate that MRHER exhibits state-of-the-art sample efficiency in benchmark tasks, outperforming RHER by 13.79% and 14.29% in the FetchPush-v1 environment and FetchPickandPlace-v1 environment, respectively." }, "2307.07768.pdf": { "id": "2307.07768v2", "title": "SoccerKDNet: A Knowledge Distillation Framework for Action Recognition in Soccer Videos", "year": "2023", "authors": [ "Sarosij Bose", "Saikat Sarkar", "Amlan Chakrabarti" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2307.07768v2", "abstract": "Classifying player actions from soccer videos is a challenging problem, which has become increasingly important in sports analytics over the years. Most state-of-the-art methods employ highly complex offline networks, which makes it difficult to deploy such models in resource constrained scenarios. Here, in this paper we propose a novel end-to-end knowledge distillation based transfer learning network pre-trained on the Kinetics400 dataset and then perform extensive analysis on the learned framework by introducing a unique loss parameterization. We also introduce a new dataset named SoccerDB1 containing 448 videos and consisting of 4 diverse classes each of players playing soccer. Furthermore, we introduce an unique loss parameter that help us linearly weigh the extent to which the predictions of each network are utilized. Finally, we also perform a thorough performance study using various changed hyperparameters. We also benchmark the first classification results on the new SoccerDB1 dataset obtaining 67.20% validation accuracy. Apart from outperforming prior arts significantly, our model also generalizes to new datasets easily. The dataset has been made publicly available at: https://bit.ly/soccerdb1" }, "2307.14732.pdf": { "id": "2307.14732v1", "title": "A Strategic Framework for Optimal Decisions in Football 1-vs-1 Shot-Taking Situations: An Integrated Approach of Machine Learning, Theory-Based Modeling, and Game Theory", "year": "2023", "authors": [ "Calvin C. K. Yeung", "Keisuke Fujii" ], "categories": "cs.LG, cs.GT", "pdf_url": "https://arxiv.org/pdf/2307.14732v1", "abstract": "Complex interactions between two opposing agents frequently occur in domains of machine learning, game theory, and other application domains. Quantitatively analyzing the strategies involved can provide an objective basis for decision-making. One such critical scenario is shot-taking in football, where decisions, such as whether the attacker should shoot or pass the ball and whether the defender should attempt to block the shot, play a crucial role in the outcome of the game. However, there are currently no effective data-driven and/or theory-based approaches to analyzing such situations. To address this issue, we proposed a novel framework to analyze such scenarios based on game theory, where we estimate the expected payoff with machine learning (ML) models, and additional features for ML models were extracted with a theory-based shot block model. Conventionally, successes or failures (1 or 0) are used as payoffs, while a success shot (goal) is extremely rare in football. Therefore, we proposed the Expected Probability of Shot On Target (xSOT) metric to evaluate players' actions even if the shot results in no goal; this allows for effective differentiation and comparison between different shots and even enables counterfactual shot situation analysis. In our experiments, we have validated the framework by comparing it with baseline and ablated models. Furthermore, we have observed a high correlation between the xSOT and existing metrics. This alignment of information suggests that xSOT provides valuable insights. Lastly, as an illustration, we studied optimal strategies in the World Cup 2022 and analyzed a shot situation in EURO 2020." }, "2307.16875.pdf": { "id": "2307.16875v1", "title": "Pyrus Base: An Open Source Python Framework for the RoboCup 2D Soccer Simulation", "year": "2023", "authors": [ "Nader Zare", "Aref Sayareh", "Omid Amini", "Mahtab Sarvmaili", "Arad Firouzkouhi", "Stan Matwin" ], "categories": "cs.RO, cs.AI", "pdf_url": "https://arxiv.org/pdf/2307.16875v1", "abstract": "Soccer, also known as football in some parts of the world, involves two teams of eleven players whose objective is to score more goals than the opposing team. To simulate this game and attract scientists from all over the world to conduct research and participate in an annual computer-based soccer world cup, Soccer Simulation 2D (SS2D) was one of the leagues initiated in the RoboCup competition. In every SS2D game, two teams of 11 players and one coach connect to the RoboCup Soccer Simulation Server and compete against each other. Over the past few years, several C++ base codes have been employed to control agents' behavior and their communication with the server. Although C++ base codes have laid the foundation for the SS2D, developing them requires an advanced level of C++ programming. C++ language complexity is a limiting disadvantage of C++ base codes for all users, especially for beginners. To conquer the challenges of C++ base codes and provide a powerful baseline for developing machine learning concepts, we introduce Pyrus, the first Python base code for SS2D. Pyrus is developed to encourage researchers to efficiently develop their ideas and integrate machine learning algorithms into their teams. Pyrus base is open-source code, and it is publicly available under MIT License on GitHub" }, "2308.01523.pdf": { "id": "2308.01523v2", "title": "Miss It Like Messi: Extracting Value from Off-Target Shots in Soccer", "year": "2023", "authors": [ "Ethan Baron", "Nathan Sandholtz", "Devin Pleuler", "Timothy C. Y. Chan" ], "categories": "stat.AP", "pdf_url": "https://arxiv.org/pdf/2308.01523v2", "abstract": "Measuring soccer shooting skill is a challenging analytics problem due to the scarcity and highly contextual nature of scoring events. The introduction of more advanced data surrounding soccer shots has given rise to model-based metrics which better cope with these challenges. Specifically, metrics such as expected goals added, goals above expectation, and post-shot expected goals all use advanced data to offer an improvement over the classical conversion rate. However, all metrics developed to date assign a value of zero to off-target shots, which account for almost two-thirds of all shots, since these shots have no probability of scoring. We posit that there is non-negligible shooting skill signal contained in the trajectories of off-target shots and propose two shooting skill metrics that incorporate the signal contained in off-target shots. Specifically, we develop a player-specific generative model for shot trajectories based on a mixture of truncated bivariate Gaussian distributions. We use this generative model to compute metrics that allow us to attach non-zero value to off-target shots. We demonstrate that our proposed metrics are more stable than current state-of-the-art metrics and have increased predictive power." }, "2308.09909.pdf": { "id": "2308.09909v1", "title": "Never Explore Repeatedly in Multi-Agent Reinforcement Learning", "year": "2023", "authors": [ "Chenghao Li", "Tonghan Wang", "Chongjie Zhang", "Qianchuan Zhao" ], "categories": "cs.LG, cs.AI, cs.MA", "pdf_url": "https://arxiv.org/pdf/2308.09909v1", "abstract": "In the realm of multi-agent reinforcement learning, intrinsic motivations have emerged as a pivotal tool for exploration. While the computation of many intrinsic rewards relies on estimating variational posteriors using neural network approximators, a notable challenge has surfaced due to the limited expressive capability of these neural statistics approximators. We pinpoint this challenge as the \"revisitation\" issue, where agents recurrently explore confined areas of the task space. To combat this, we propose a dynamic reward scaling approach. This method is crafted to stabilize the significant fluctuations in intrinsic rewards in previously explored areas and promote broader exploration, effectively curbing the revisitation phenomenon. Our experimental findings underscore the efficacy of our approach, showcasing enhanced performance in demanding environments like Google Research Football and StarCraft II micromanagement tasks, especially in sparse reward settings." }, "2308.11272.pdf": { "id": "2308.11272v2", "title": "FoX: Formation-aware exploration in multi-agent reinforcement learning", "year": "2023", "authors": [ "Yonghyeon Jo", "Sunwoo Lee", "Junghyuk Yeom", "Seungyul Han" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2308.11272v2", "abstract": "Recently, deep multi-agent reinforcement learning (MARL) has gained significant popularity due to its success in various cooperative multi-agent tasks. However, exploration still remains a challenging problem in MARL due to the partial observability of the agents and the exploration space that can grow exponentially as the number of agents increases. Firstly, in order to address the scalability issue of the exploration space, we define a formation-based equivalence relation on the exploration space and aim to reduce the search space by exploring only meaningful states in different formations. Then, we propose a novel formation-aware exploration (FoX) framework that encourages partially observable agents to visit the states in diverse formations by guiding them to be well aware of their current formation solely based on their own observations. Numerical results show that the proposed FoX framework significantly outperforms the state-of-the-art MARL algorithms on Google Research Football (GRF) and sparse Starcraft II multi-agent challenge (SMAC) tasks." }, "2308.15559.pdf": { "id": "2308.15559v1", "title": "Glocal Explanations of Expected Goal Models in Soccer", "year": "2023", "authors": [ "Mustafa Cavus", "Adrian Stando", "Przemyslaw Biecek" ], "categories": "cs.LG, stat.ML", "pdf_url": "https://arxiv.org/pdf/2308.15559v1", "abstract": "The expected goal models have gained popularity, but their interpretability is often limited, especially when trained using black-box methods. Explainable artificial intelligence tools have emerged to enhance model transparency and extract descriptive knowledge for a single observation or for all observations. However, explaining black-box models for a specific group of observations may be more useful in some domains. This paper introduces the glocal explanations (between local and global levels) of the expected goal models to enable performance analysis at the team and player levels by proposing the use of aggregated versions of the SHAP values and partial dependence profiles. This allows knowledge to be extracted from the expected goal model for a player or team rather than just a single shot. In addition, we conducted real-data applications to illustrate the usefulness of aggregated SHAP and aggregated profiles. The paper concludes with remarks on the potential of these explanations for performance analysis in soccer analytics." }, "2309.01526.pdf": { "id": "2309.01526v1", "title": "Passing Heatmap Prediction Based on Transformer Model and Tracking Data", "year": "2023", "authors": [ "Yisheng Pei", "Varuna De Silva", "Mike Caine" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2309.01526v1", "abstract": "Although the data-driven analysis of football players' performance has been developed for years, most research only focuses on the on-ball event including shots and passes, while the off-ball movement remains a little-explored area in this domain. Players' contributions to the whole match are evaluated unfairly, those who have more chances to score goals earn more credit than others, while the indirect and unnoticeable impact that comes from continuous movement has been ignored. This research presents a novel deep-learning network architecture which is capable to predict the potential end location of passes and how players' movement before the pass affects the final outcome. Once analysed more than 28,000 pass events, a robust prediction can be achieved with more than 0.7 Top-1 accuracy. And based on the prediction, a better understanding of the pitch control and pass option could be reached to measure players' off-ball movement contribution to defensive performance. Moreover, this model could provide football analysts a better tool and metric to understand how players' movement over time contributes to the game strategy and final victory." }, "2309.06285.pdf": { "id": "2309.06285v1", "title": "Jersey Number Recognition using Keyframe Identification from Low-Resolution Broadcast Videos", "year": "2023", "authors": [ "Bavesh Balaji", "Jerrin Bright", "Harish Prakash", "Yuhao Chen", "David A Clausi", "John Zelek" ], "categories": "cs.CV, cs.AI", "pdf_url": "https://arxiv.org/pdf/2309.06285v1", "abstract": "Player identification is a crucial component in vision-driven soccer analytics, enabling various downstream tasks such as player assessment, in-game analysis, and broadcast production. However, automatically detecting jersey numbers from player tracklets in videos presents challenges due to motion blur, low resolution, distortions, and occlusions. Existing methods, utilizing Spatial Transformer Networks, CNNs, and Vision Transformers, have shown success in image data but struggle with real-world video data, where jersey numbers are not visible in most of the frames. Hence, identifying frames that contain the jersey number is a key sub-problem to tackle. To address these issues, we propose a robust keyframe identification module that extracts frames containing essential high-level information about the jersey number. A spatio-temporal network is then employed to model spatial and temporal context and predict the probabilities of jersey numbers in the video. Additionally, we adopt a multi-task loss function to predict the probability distribution of each digit separately. Extensive evaluations on the SoccerNet dataset demonstrate that incorporating our proposed keyframe identification module results in a significant 37.81% and 37.70% increase in the accuracies of 2 different test sets with domain gaps. These results highlight the effectiveness and importance of our approach in tackling the challenges of automatic jersey number detection in sports videos." }, "2309.06802.pdf": { "id": "2309.06802v1", "title": "Dynamic NeRFs for Soccer Scenes", "year": "2023", "authors": [ "Sacha Lewin", "Maxime Vandegar", "Thomas Hoyoux", "Olivier Barnich", "Gilles Louppe" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2309.06802v1", "abstract": "The long-standing problem of novel view synthesis has many applications, notably in sports broadcasting. Photorealistic novel view synthesis of soccer actions, in particular, is of enormous interest to the broadcast industry. Yet only a few industrial solutions have been proposed, and even fewer that achieve near-broadcast quality of the synthetic replays. Except for their setup of multiple static cameras around the playfield, the best proprietary systems disclose close to no information about their inner workings. Leveraging multiple static cameras for such a task indeed presents a challenge rarely tackled in the literature, for a lack of public datasets: the reconstruction of a large-scale, mostly static environment, with small, fast-moving elements. Recently, the emergence of neural radiance fields has induced stunning progress in many novel view synthesis applications, leveraging deep learning principles to produce photorealistic results in the most challenging settings. In this work, we investigate the feasibility of basing a solution to the task on dynamic NeRFs, i.e., neural models purposed to reconstruct general dynamic content. We compose synthetic soccer environments and conduct multiple experiments using them, identifying key components that help reconstruct soccer scenes with dynamic NeRFs. We show that, although this approach cannot fully meet the quality requirements for the target application, it suggests promising avenues toward a cost-efficient, automatic solution. We also make our work dataset and code publicly available, with the goal to encourage further efforts from the research community on the task of novel view synthesis for dynamic soccer scenes. For code, data, and video results, please see https://soccernerfs.isach.be." }, "2309.12067.pdf": { "id": "2309.12067v1", "title": "Survey of Action Recognition, Spotting and Spatio-Temporal Localization in Soccer -- Current Trends and Research Perspectives", "year": "2023", "authors": [ "Karolina Seweryn", "Anna Wr\u00f3blewska", "Szymon \u0141ukasik" ], "categories": "cs.CV, cs.AI, cs.LG", "pdf_url": "https://arxiv.org/pdf/2309.12067v1", "abstract": "Action scene understanding in soccer is a challenging task due to the complex and dynamic nature of the game, as well as the interactions between players. This article provides a comprehensive overview of this task divided into action recognition, spotting, and spatio-temporal action localization, with a particular emphasis on the modalities used and multimodal methods. We explore the publicly available data sources and metrics used to evaluate models' performance. The article reviews recent state-of-the-art methods that leverage deep learning techniques and traditional methods. We focus on multimodal methods, which integrate information from multiple sources, such as video and audio data, and also those that represent one source in various ways. The advantages and limitations of methods are discussed, along with their potential for improving the accuracy and robustness of models. Finally, the article highlights some of the open research questions and future directions in the field of soccer action recognition, including the potential for multimodal methods to advance this field. Overall, this survey provides a valuable resource for researchers interested in the field of action scene understanding in soccer." }, "2309.12951.pdf": { "id": "2309.12951v1", "title": "Boosting Studies of Multi-Agent Reinforcement Learning on Google Research Football Environment: the Past, Present, and Future", "year": "2023", "authors": [ "Yan Song", "He Jiang", "Haifeng Zhang", "Zheng Tian", "Weinan Zhang", "Jun Wang" ], "categories": "cs.MA", "pdf_url": "https://arxiv.org/pdf/2309.12951v1", "abstract": "Even though Google Research Football (GRF) was initially benchmarked and studied as a single-agent environment in its original paper, recent years have witnessed an increasing focus on its multi-agent nature by researchers utilizing it as a testbed for Multi-Agent Reinforcement Learning (MARL). However, the absence of standardized environment settings and unified evaluation metrics for multi-agent scenarios hampers the consistent understanding of various studies. Furthermore, the challenging 5-vs-5 and 11-vs-11 full-game scenarios have received limited thorough examination due to their substantial training complexities. To address these gaps, this paper extends the original environment by not only standardizing the environment settings and benchmarking cooperative learning algorithms across different scenarios, including the most challenging full-game scenarios, but also by discussing approaches to enhance football AI from diverse perspectives and introducing related research tools. Specifically, we provide a distributed and asynchronous population-based self-play framework with diverse pre-trained policies for faster training, two football-specific analytical tools for deeper investigation, and an online leaderboard for broader evaluation. The overall expectation of this work is to advance the study of Multi-Agent Reinforcement Learning on Google Research Football environment, with the ultimate goal of benefiting real-world sports beyond virtual games." }, "2309.14807.pdf": { "id": "2309.14807v1", "title": "Evaluating Soccer Match Prediction Models: A Deep Learning Approach and Feature Optimization for Gradient-Boosted Trees", "year": "2023", "authors": [ "Calvin Yeung", "Rory Bunker", "Rikuhei Umemoto", "Keisuke Fujii" ], "categories": "cs.LG, cs.AI", "pdf_url": "https://arxiv.org/pdf/2309.14807v1", "abstract": "Machine learning models have become increasingly popular for predicting the results of soccer matches, however, the lack of publicly-available benchmark datasets has made model evaluation challenging. The 2023 Soccer Prediction Challenge required the prediction of match results first in terms of the exact goals scored by each team, and second, in terms of the probabilities for a win, draw, and loss. The original training set of matches and features, which was provided for the competition, was augmented with additional matches that were played between 4 April and 13 April 2023, representing the period after which the training set ended, but prior to the first matches that were to be predicted (upon which the performance was evaluated). A CatBoost model was employed using pi-ratings as the features, which were initially identified as the optimal choice for calculating the win/draw/loss probabilities. Notably, deep learning models have frequently been disregarded in this particular task. Therefore, in this study, we aimed to assess the performance of a deep learning model and determine the optimal feature set for a gradient-boosted tree model. The model was trained using the most recent five years of data, and three training and validation sets were used in a hyperparameter grid search. The results from the validation sets show that our model had strong performance and stability compared to previously published models from the 2017 Soccer Prediction Challenge for win/draw/loss prediction." }, "2310.04796.pdf": { "id": "2310.04796v3", "title": "Accelerate Multi-Agent Reinforcement Learning in Zero-Sum Games with Subgame Curriculum Learning", "year": "2023", "authors": [ "Jiayu Chen", "Zelai Xu", "Yunfei Li", "Chao Yu", "Jiaming Song", "Huazhong Yang" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2310.04796v3", "abstract": "Learning Nash equilibrium (NE) in complex zero-sum games with multi-agent reinforcement learning (MARL) can be extremely computationally expensive. Curriculum learning is an effective way to accelerate learning, but an under-explored dimension for generating a curriculum is the difficulty-to-learn of the subgames -- games induced by starting from a specific state. In this work, we present a novel subgame curriculum learning framework for zero-sum games. It adopts an adaptive initial state distribution by resetting agents to some previously visited states where they can quickly learn to improve performance. Building upon this framework, we derive a subgame selection metric that approximates the squared distance to NE values and further adopt a particle-based state sampler for subgame generation. Integrating these techniques leads to our new algorithm, Subgame Automatic Curriculum Learning (SACL), which is a realization of the subgame curriculum learning framework. SACL can be combined with any MARL algorithm such as MAPPO. Experiments in the particle-world environment and Google Research Football environment show SACL produces much stronger policies than baselines. In the challenging hide-and-seek quadrant environment, SACL produces all four emergent stages and uses only half the samples of MAPPO with self-play. The project website is at https://sites.google.com/view/sacl-rl." }, "2310.04797.pdf": { "id": "2310.04797v1", "title": "Penalty shootouts are tough, but the alternating order is fair", "year": "2023", "authors": [ "Silvan Vollmer", "David Schoch", "Ulrik Brandes" ], "categories": "stat.AP", "pdf_url": "https://arxiv.org/pdf/2310.04797v1", "abstract": "We compare conversion rates of association football (soccer) penalties during regulation or extra time with those during shootouts. Our data consists of roughly 50,000 penalties from the eleven~most recent seasons in European men's football competitions. About one third of the penalties are from more than 1,500 penalty shootouts. We find that shootout conversion rates are significantly lower, and attribute this to worse performance of shooters rather than better performance of goalkeepers. We also find that, statistically, there is no advantage for either team in the usual alternating shooting order. These main findings are complemented by a number of more detailed analyses." }, "2310.04912.pdf": { "id": "2310.04912v2", "title": "Intelligent Sampling Consensus for Homography Estimation in Football Videos Using Featureless Unpaired Points", "year": "2023", "authors": [ "George Nousias", "Konstantinos Delibasis", "Ilias Maglogiannis" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2310.04912v2", "abstract": "Estimating the homography matrix between images captured under radically different camera poses and zoom factors is a complex challenge. Traditional methods rely on the Random Sample Consensus (RANSAC) algorithm, which requires pairs of homologous points, pre-matched based on local image feature vectors. Sampling consensus is a core step in many Artificial Intelligence (AI) algorithms that enable computer systems to recognize patterns in data. In this paper, we propose H-RANSAC, an algorithm for homography estimation that eliminates the need for feature vectors or explicit point pairing, while it optionally supports point labeling into two classes. H-RANSAC introduces a novel geometric (cheiral) criterion that intelligently rejects implausible point configurations at the beginning of each iteration, while leveraging concave quadrilaterals typically discarded by similar algorithms. A post-hoc criterion at the end of each iteration improves accuracy further. Analytical derivations of the expected maximum iterations are provided, considering success probabilities and outlier rates, enabling adaptive performance tuning. The algorithm is validated on a demanding task: estimating homography between video frames of football matches captured by 12 cameras with highly divergent viewpoints. Results show that H-RANSAC significantly outperforms state-of-the-art classical methods, combined with deep learning-based salient point detection, in terms of average reprojection error and success rates. The relevant implementation is available in https://github.com/gnousias/H-RANSAC." }, "2310.10553.pdf": { "id": "2310.10553v2", "title": "TacticAI: an AI assistant for football tactics", "year": "2023", "authors": [ "Zhe Wang", "Petar Veli\u010dkovi\u0107", "Daniel Hennes", "Nenad Toma\u0161ev", "Laurel Prince", "Michael Kaisers" ], "categories": "cs.LG, cs.MA, stat.ML", "pdf_url": "https://arxiv.org/pdf/2310.10553v2", "abstract": "Identifying key patterns of tactics implemented by rival teams, and developing effective responses, lies at the heart of modern football. However, doing so algorithmically remains an open research challenge. To address this unmet need, we propose TacticAI, an AI football tactics assistant developed and evaluated in close collaboration with domain experts from Liverpool FC. We focus on analysing corner kicks, as they offer coaches the most direct opportunities for interventions and improvements. TacticAI incorporates both a predictive and a generative component, allowing the coaches to effectively sample and explore alternative player setups for each corner kick routine and to select those with the highest predicted likelihood of success. We validate TacticAI on a number of relevant benchmark tasks: predicting receivers and shot attempts and recommending player position adjustments. The utility of TacticAI is validated by a qualitative study conducted with football domain experts at Liverpool FC. We show that TacticAI's model suggestions are not only indistinguishable from real tactics, but also favoured over existing tactics 90% of the time, and that TacticAI offers an effective corner kick retrieval system. TacticAI achieves these results despite the limited availability of gold-standard data, achieving data efficiency through geometric deep learning." }, "2310.12328.pdf": { "id": "2310.12328v1", "title": "Soccer on Social Media", "year": "2023", "authors": [ "Mehdi Houshmand Sarkhoosh", "Sayed Mohammad Majidi Dorcheh", "Sushant Gautam", "Cise Midoglu", "Saeed Shafiee Sabet", "P\u00e5l Halvorsen" ], "categories": "cs.MM, cs.SI", "pdf_url": "https://arxiv.org/pdf/2310.12328v1", "abstract": "In the era of digitalization, social media has become an integral part of our lives, serving as a significant hub for individuals and businesses to share information, communicate, and engage. This is also the case for professional sports, where leagues, clubs and players are using social media to reach out to their fans. In this respect, a huge amount of time is spent curating multimedia content for various social media platforms and their target users. With the emergence of Artificial Intelligence (AI), AI-based tools for automating content generation and enhancing user experiences on social media have become widely popular. However, to effectively utilize such tools, it is imperative to comprehend the demographics and preferences of users on different platforms, understand how content providers post information in these channels, and how different types of multimedia are consumed by audiences. This report presents an analysis of social media platforms, in terms of demographics, supported multimedia modalities, and distinct features and specifications for different modalities, followed by a comparative case study of select European soccer leagues and teams, in terms of their social media practices. Through this analysis, we demonstrate that social media, while being very important for and widely used by supporters from all ages, also requires a fine-tuned effort on the part of soccer professionals, in order to elevate fan experiences and foster engagement." }, "2310.14553.pdf": { "id": "2310.14553v1", "title": "Denoising Opponents Position in Partial Observation Environment", "year": "2023", "authors": [ "Aref Sayareh", "Aria Sardari", "Vahid Khoddami", "Nader Zare", "Vinicius Prado da Fonseca", "Amilcar Soares" ], "categories": "cs.RO, cs.AI, cs.MA", "pdf_url": "https://arxiv.org/pdf/2310.14553v1", "abstract": "The RoboCup competitions hold various leagues, and the Soccer Simulation 2D League is a major among them. Soccer Simulation 2D (SS2D) match involves two teams, including 11 players and a coach for each team, competing against each other. The players can only communicate with the Soccer Simulation Server during the game. Several code bases are released publicly to simplify team development. So researchers can easily focus on decision-making and implementing machine learning methods. SS2D actions and behaviors are only partially accurate due to different challenges, such as noise and partial observation. Therefore, one strategy is to implement alternative denoising methods to tackle observation inaccuracy. Our idea is to predict opponent positions while they have yet to be seen in a finite number of cycles using machine learning methods to make more accurate actions such as pass. We will explain our position prediction idea powered by Long Short-Term Memory models (LSTM) and Deep Neural Networks (DNN). The results show that the LSTM and DNN predict the opponents' position more accurately than the standard algorithm, such as the last-seen method." }, "2310.16368.pdf": { "id": "2310.16368v1", "title": "Transformer-based Live Update Generation for Soccer Matches from Microblog Posts", "year": "2023", "authors": [ "Masashi Oshika", "Kosuke Yamada", "Ryohei Sasano", "Koichi Takeda" ], "categories": "cs.CL", "pdf_url": "https://arxiv.org/pdf/2310.16368v1", "abstract": "It has been known to be difficult to generate adequate sports updates from a sequence of vast amounts of diverse live tweets, although the live sports viewing experience with tweets is gaining the popularity. In this paper, we focus on soccer matches and work on building a system to generate live updates for soccer matches from tweets so that users can instantly grasp a match's progress and enjoy the excitement of the match from raw tweets. Our proposed system is based on a large pre-trained language model and incorporates a mechanism to control the number of updates and a mechanism to reduce the redundancy of duplicate and similar updates." }, "2311.02097.pdf": { "id": "2311.02097v1", "title": "The CO2 Footprint of Club Soccer in the US", "year": "2023", "authors": [ "Aarya Upadhyay" ], "categories": "physics.soc-ph", "pdf_url": "https://arxiv.org/pdf/2311.02097v1", "abstract": "I played youth soccer for 11 years starting age at 5. As I progressed through the age groups and got better at the game, I moved up in club levels under the US Club Soccer system. It was during my time as an ECNL player that I began to realize the magnitude of the travel burden associated with local, regional and out of state games. Most of the travel was by road and team members often travelled with minimal car-pooling. Just about that time I was also becoming aware of (and interested in) Climate Change and the various factors impacting climate change including Greenhouse Gases. It was during these travels that I began to track fuel consumption and the idea of estimating the CO2 impact of these soccer road trips was born. In this paper I report my findings through simple calculations and provide extrapolations through simulated scenario experiments. Based on my findings I discuss a few potential solutions that may be considered to reduce the CO2 footprint associated with US youth travel soccer. The idea behind this paper is to raise awareness amongst soccer families and to engage my fellow soccer peers into making the sport we all love as green as possible." }, "2311.04599.pdf": { "id": "2311.04599v2", "title": "Explainable artificial intelligence model for identifying Market Value in Professional Soccer Players", "year": "2023", "authors": [ "Chunyang Huang", "Shaoliang Zhang" ], "categories": "cs.LG, cs.CY, q-fin.CP", "pdf_url": "https://arxiv.org/pdf/2311.04599v2", "abstract": "This study introduces an advanced machine learning method for predicting soccer players' market values, combining ensemble models and the Shapley Additive Explanations (SHAP) for interpretability. Utilizing data from about 12,000 players from Sofifa, the Boruta algorithm streamlined feature selection. The Gradient Boosting Decision Tree (GBDT) model excelled in predictive accuracy, with an R-squared of 0.901 and a Root Mean Squared Error (RMSE) of 3,221,632.175. Player attributes in skills, fitness, and cognitive areas significantly influenced market value. These insights aid sports industry stakeholders in player valuation. However, the study has limitations, like underestimating superstar players' values and needing larger datasets. Future research directions include enhancing the model's applicability and exploring value prediction in various contexts." }, "2311.06892.pdf": { "id": "2311.06892v1", "title": "Setting a Baseline for long-shot real-time Player and Ball detection in Soccer Videos", "year": "2023", "authors": [ "Konstantinos Moutselos", "Ilias Maglogiannis" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2311.06892v1", "abstract": "Players and ball detection are among the first required steps on a football analytics platform. Until recently, the existing open datasets on which the evaluations of most models were based, were not sufficient. In this work, we point out their weaknesses, and with the advent of the SoccerNet v3, we propose and deliver to the community an edited part of its dataset, in YOLO normalized annotation format for training and evaluation. The code of the methods and metrics are provided so that they can be used as a benchmark in future comparisons. The recent YOLO8n model proves better than FootAndBall in long-shot real-time detection of the ball and players on football fields." }, "2311.13707.pdf": { "id": "2311.13707v1", "title": "Bayes-xG: Player and Position Correction on Expected Goals (xG) using Bayesian Hierarchical Approach", "year": "2023", "authors": [ "Alexander Scholtes", "Oktay Karaku\u015f" ], "categories": "stat.AP, cs.LG", "pdf_url": "https://arxiv.org/pdf/2311.13707v1", "abstract": "This study employs Bayesian methodologies to explore the influence of player or positional factors in predicting the probability of a shot resulting in a goal, measured by the expected goals (xG) metric. Utilising publicly available data from StatsBomb, Bayesian hierarchical logistic regressions are constructed, analysing approximately 10,000 shots from the English Premier League to ascertain whether positional or player-level effects impact xG. The findings reveal positional effects in a basic model that includes only distance to goal and shot angle as predictors, highlighting that strikers and attacking midfielders exhibit a higher likelihood of scoring. However, these effects diminish when more informative predictors are introduced. Nevertheless, even with additional predictors, player-level effects persist, indicating that certain players possess notable positive or negative xG adjustments, influencing their likelihood of scoring a given chance. The study extends its analysis to data from Spain's La Liga and Germany's Bundesliga, yielding comparable results. Additionally, the paper assesses the impact of prior distribution choices on outcomes, concluding that the priors employed in the models provide sound results but could be refined to enhance sampling efficiency for constructing more complex and extensive models feasibly." }, "2311.14642.pdf": { "id": "2311.14642v1", "title": "Continuous football player tracking from discrete broadcast data", "year": "2023", "authors": [ "Matthew J. Penn", "Christl A. Donnelly", "Samir Bhatt" ], "categories": "cs.CV, cs.MA", "pdf_url": "https://arxiv.org/pdf/2311.14642v1", "abstract": "Player tracking data remains out of reach for many professional football teams as their video feeds are not sufficiently high quality for computer vision technologies to be used. To help bridge this gap, we present a method that can estimate continuous full-pitch tracking data from discrete data made from broadcast footage. Such data could be collected by clubs or players at a similar cost to event data, which is widely available down to semi-professional level. We test our method using open-source tracking data, and include a version that can be applied to a large set of over 200 games with such discrete data." }, "2312.04338.pdf": { "id": "2312.04338v1", "title": "Stochastic modelling of football matches", "year": "2023", "authors": [ "Luiz Fernando G. N. Maia", "Teemu Pennanen", "Moacyr A. H. B. da Silva", "Rodrigo S. Targino" ], "categories": "stat.AP", "pdf_url": "https://arxiv.org/pdf/2312.04338v1", "abstract": "This paper develops a general framework for stochastic modeling of goals and other events in football (soccer) matches. The events are modelled as Cox processes (doubly stochastic Poisson processes) where the event intensities may depend on all the modeled events as well as external factors. The model has a strictly concave log-likelihood function which facilitates its fitting to observed data. Besides event times, the model describes the random lengths of stoppage times which can have a strong influence on the final score of a match. The model is illustrated on eight years of data from Campeonato Brasileiro de Futebol S\u00e9rie A. We find that dynamic regressors significantly improve the in-game predictive power of the model. In particular, a) when a team receives a red card, its goal intensity decreases more than 30%; b) the goal rate of a team increases by 10% if it is losing by one goal and by 20% if its losing by two goals; and c) when the goal difference at the end of the second half is less than or equal to one, the stoppage time is on average more than one minute longer than in matches with a difference of two goals." }, "2312.05783.pdf": { "id": "2312.05783v1", "title": "DCIR: Dynamic Consistency Intrinsic Reward for Multi-Agent Reinforcement Learning", "year": "2023", "authors": [ "Kunyang Lin", "Yufeng Wang", "Peihao Chen", "Runhao Zeng", "Siyuan Zhou", "Mingkui Tan" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2312.05783v1", "abstract": "Learning optimal behavior policy for each agent in multi-agent systems is an essential yet difficult problem. Despite fruitful progress in multi-agent reinforcement learning, the challenge of addressing the dynamics of whether two agents should exhibit consistent behaviors is still under-explored. In this paper, we propose a new approach that enables agents to learn whether their behaviors should be consistent with that of other agents by utilizing intrinsic rewards to learn the optimal policy for each agent. We begin by defining behavior consistency as the divergence in output actions between two agents when provided with the same observation. Subsequently, we introduce dynamic consistency intrinsic reward (DCIR) to stimulate agents to be aware of others' behaviors and determine whether to be consistent with them. Lastly, we devise a dynamic scale network (DSN) that provides learnable scale factors for the agent at every time step to dynamically ascertain whether to award consistent behavior and the magnitude of rewards. We evaluate DCIR in multiple environments including Multi-agent Particle, Google Research Football and StarCraft II Micromanagement, demonstrating its efficacy." }, "2312.12700.pdf": { "id": "2312.12700v1", "title": "Performance rating in chess, tennis, and other contexts", "year": "2023", "authors": [ "Mehmet S. Ismail" ], "categories": "econ.TH, cs.MA", "pdf_url": "https://arxiv.org/pdf/2312.12700v1", "abstract": "In this note, I introduce Estimated Performance Rating (PR$^e$), a novel system for evaluating player performance in sports and games. PR$^e$ addresses a key limitation of the Tournament Performance Rating (TPR) system, which is undefined for zero or perfect scores in a series of games. PR$^e$ is defined as the rating that solves an optimization problem related to scoring probability, making it applicable for any performance level. The main theorem establishes that the PR$^e$ of a player is equivalent to the TPR whenever the latter is defined. I then apply this system to historically significant win-streaks in association football, tennis, and chess. Beyond sports, PR$^e$ has broad applicability in domains where Elo ratings are used, from college rankings to the evaluation of large language models." }, "2312.15236.pdf": { "id": "2312.15236v1", "title": "Classifying Soccer Ball-on-Goal Position Through Kicker Shooting Action", "year": "2023", "authors": [ "Javier Tor\u00f3n-Artiles", "Daniel Hern\u00e1ndez-Sosa", "Oliverio J. Santana", "Javier Lorenzo-Navarro", "David Freire-Obreg\u00f3n" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2312.15236v1", "abstract": "This research addresses whether the ball's direction after a soccer free-kick can be accurately predicted solely by observing the shooter's kicking technique. To investigate this, we meticulously curated a dataset of soccer players executing free kicks and conducted manual temporal segmentation to identify the moment of the kick precisely. Our approach involves utilizing neural networks to develop a model that integrates Human Action Recognition (HAR) embeddings with contextual information, predicting the ball-on-goal position (BoGP) based on two temporal states: the kicker's run-up and the instant of the kick. The study encompasses a performance evaluation for eleven distinct HAR backbones, shedding light on their effectiveness in BoGP estimation during free-kick situations. An extra tabular metadata input is introduced, leading to an interesting model enhancement without introducing bias. The promising results reveal 69.1% accuracy when considering two primary BoGP classes: right and left. This underscores the model's proficiency in predicting the ball's destination towards the goal with high accuracy, offering promising implications for understanding free-kick dynamics in soccer." }, "2312.16273.pdf": { "id": "2312.16273v1", "title": "Coordination and Machine Learning in Multi-Robot Systems: Applications in Robotic Soccer", "year": "2023", "authors": [ "Luis Paulo Reis" ], "categories": "cs.RO", "pdf_url": "https://arxiv.org/pdf/2312.16273v1", "abstract": "This paper presents the concepts of Artificial Intelligence, Multi-Agent-Systems, Coordination, Intelligent Robotics and Deep Reinforcement Learning. Emphasis is given on and how AI and DRL, may be efficiently used to create efficient robot skills and coordinated robotic teams, capable of performing very complex actions and tasks, such as playing a game of soccer. The paper also presents the concept of robotic soccer and the vision and structure of the RoboCup initiative with emphasis on the Humanoid Simulation 3D league and the new challenges this competition, poses. The final topics presented at the paper are based on the research developed/coordinated by the author throughout the last 22 years in the context of the FCPortugal project. The paper presents a short description of the coordination methodologies developed, such as: Strategy, Tactics, Formations, Setplays, and Coaching Languages and the use of Machine Learning to optimize the use of this concepts. The topics presented also include novel stochastic search algorithms for black box optimization and their use in the optimization of omnidirectional walking skills, robotic multi-agent learning and the creation of a humanoid kick with controlled distance. Finally, new applications using variations of the Proximal Policy Optimization algorithm and advanced modelling for robot and multi-robot learning are briefly explained with emphasis for our new humanoid sprinting and running skills and an amazing humanoid robot soccer dribbling skill. FCPortugal project enabled us to publish more than 100 papers and win several competitions in different leagues and many scientific awards at RoboCup. In total, our team won more than 40 awards in international competitions including a clear victory at the Simulation 3D League at RoboCup 2022 competition, scoring 84 goals and conceding only 2." }, "2401.03406.pdf": { "id": "2401.03406v1", "title": "Improving Dribbling, Passing, and Marking Actions in Soccer Simulation 2D Games Using Machine Learning", "year": "2024", "authors": [ "Nader Zare", "Omid Amini", "Aref Sayareh", "Mahtab Sarvmaili", "Arad Firouzkouhi", "Stan Matwin" ], "categories": "cs.RO, cs.AI, cs.LG", "pdf_url": "https://arxiv.org/pdf/2401.03406v1", "abstract": "The RoboCup competition was started in 1997, and is known as the oldest RoboCup league. The RoboCup 2D Soccer Simulation League is a stochastic, partially observable soccer environment in which 24 autonomous agents play on two opposing teams. In this paper, we detail the main strategies and functionalities of CYRUS, the RoboCup 2021 2D Soccer Simulation League champions. The new functionalities presented and discussed in this work are (i) Multi Action Dribble, (ii) Pass Prediction and (iii) Marking Decision. The Multi Action Dribbling strategy enabled CYRUS to succeed more often and to be safer when dribbling actions were performed during a game. The Pass Prediction enhanced our gameplay by predicting our teammate's passing behavior, anticipating and making our agents collaborate better towards scoring goals. Finally, the Marking Decision addressed the multi-agent matching problem to improve CYRUS defensive strategy by finding an optimal solution to mark opponents' players." }, "2401.03410.pdf": { "id": "2401.03410v1", "title": "Engineering Features to Improve Pass Prediction in Soccer Simulation 2D Games", "year": "2024", "authors": [ "Nader Zare", "Mahtab Sarvmaili", "Aref Sayareh", "Omid Amini", "Stan Matwin Amilcar Soares" ], "categories": "cs.AI, cs.LG, cs.RO", "pdf_url": "https://arxiv.org/pdf/2401.03410v1", "abstract": "Soccer Simulation 2D (SS2D) is a simulation of a real soccer game in two dimensions. In soccer, passing behavior is an essential action for keeping the ball in possession of our team and creating goal opportunities. Similarly, for SS2D, predicting the passing behaviors of both opponents and our teammates helps manage resources and score more goals. Therefore, in this research, we have tried to address the modeling of passing behavior of soccer 2D players using Deep Neural Networks (DNN) and Random Forest (RF). We propose an embedded data extraction module that can record the decision-making of agents in an online format. Afterward, we apply four data sorting techniques for training data preparation. After, we evaluate the trained models' performance playing against 6 top teams of RoboCup 2019 that have distinctive playing strategies. Finally, we examine the importance of different feature groups on the prediction of a passing strategy. All results in each step of this work prove our suggested methodology's effectiveness and improve the performance of the pass prediction in Soccer Simulation 2D games ranging from 5\\% (e.g., playing against the same team) to 10\\% (e.g., playing against Robocup top teams)." }, "2401.08718.pdf": { "id": "2401.08718v1", "title": "Investigating Fouling Efficiency in Football Using Expected Booking (xB) Model", "year": "2024", "authors": [ "Adnan Azmat", "Su Su Yi" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2401.08718v1", "abstract": "This paper introduces the Expected Booking (xB) model, a novel metric designed to estimate the likelihood of a foul resulting in a yellow card in football. Through three iterative experiments, employing ensemble methods, the model demonstrates improved performance with additional features and an expanded dataset. Analysis of FIFA World Cup 2022 data validates the model's efficacy in providing insights into team and player fouling tactics, aligning with actual defensive performance. The xB model addresses a gap in fouling efficiency examination, emphasizing defensive strategies which often overlooked. Further enhancements are suggested through the incorporation of comprehensive data and spatial features." }, "2401.09940.pdf": { "id": "2401.09940v2", "title": "Biases in Expected Goals Models Confound Finishing Ability", "year": "2024", "authors": [ "Jesse Davis", "Pieter Robberechts" ], "categories": "cs.LG, stat.AP", "pdf_url": "https://arxiv.org/pdf/2401.09940v2", "abstract": "Expected Goals (xG) has emerged as a popular tool for evaluating finishing skill in soccer analytics. It involves comparing a player's cumulative xG with their actual goal output, where consistent overperformance indicates strong finishing ability. However, the assessment of finishing skill in soccer using xG remains contentious due to players' difficulty in consistently outperforming their cumulative xG. In this paper, we aim to address the limitations and nuances surrounding the evaluation of finishing skill using xG statistics. Specifically, we explore three hypotheses: (1) the deviation between actual and expected goals is an inadequate metric due to the high variance of shot outcomes and limited sample sizes, (2) the inclusion of all shots in cumulative xG calculation may be inappropriate, and (3) xG models contain biases arising from interdependencies in the data that affect skill measurement. We found that sustained overperformance of cumulative xG requires both high shot volumes and exceptional finishing, including all shot types can obscure the finishing ability of proficient strikers, and that there is a persistent bias that makes the actual and expected goals closer for excellent finishers than it really is. Overall, our analysis indicates that we need more nuanced quantitative approaches for investigating a player's finishing ability, which we achieved using a technique from AI fairness to learn an xG model that is calibrated for multiple subgroups of players. As a concrete use case, we show that (1) the standard biased xG model underestimates Messi's GAX by 17% and (2) Messi's GAX is 27% higher than the typical elite high-shot-volume attacker, indicating that Messi is even a more exceptional finisher than people commonly believed." }, "2401.13460.pdf": { "id": "2401.13460v3", "title": "Multi-Agent Diagnostics for Robustness via Illuminated Diversity", "year": "2024", "authors": [ "Mikayel Samvelyan", "Davide Paglieri", "Minqi Jiang", "Jack Parker-Holder", "Tim Rockt\u00e4schel" ], "categories": "cs.LG, cs.AI, cs.MA", "pdf_url": "https://arxiv.org/pdf/2401.13460v3", "abstract": "In the rapidly advancing field of multi-agent systems, ensuring robustness in unfamiliar and adversarial settings is crucial. Notwithstanding their outstanding performance in familiar environments, these systems often falter in new situations due to overfitting during the training phase. This is especially pronounced in settings where both cooperative and competitive behaviours are present, encapsulating a dual nature of overfitting and generalisation challenges. To address this issue, we present Multi-Agent Diagnostics for Robustness via Illuminated Diversity (MADRID), a novel approach for generating diverse adversarial scenarios that expose strategic vulnerabilities in pre-trained multi-agent policies. Leveraging the concepts from open-ended learning, MADRID navigates the vast space of adversarial settings, employing a target policy's regret to gauge the vulnerabilities of these settings. We evaluate the effectiveness of MADRID on the 11vs11 version of Google Research Football, one of the most complex environments for multi-agent reinforcement learning. Specifically, we employ MADRID for generating a diverse array of adversarial settings for TiZero, the state-of-the-art approach which \"masters\" the game through 45 days of training on a large-scale distributed infrastructure. We expose key shortcomings in TiZero's tactical decision-making, underlining the crucial importance of rigorous evaluation in multi-agent systems." }, "2401.16795.pdf": { "id": "2401.16795v1", "title": "Performance Insights-based AI-driven Football Transfer Fee Prediction", "year": "2024", "authors": [ "Daniil Sulimov" ], "categories": "cs.LG, cs.AI", "pdf_url": "https://arxiv.org/pdf/2401.16795v1", "abstract": "We developed an artificial intelligence approach to predict the transfer fee of a football player. This model can help clubs make better decisions about which players to buy and sell, which can lead to improved performance and increased club budgets. Having collected data on player performance, transfer fees, and other factors that might affect a player's value, we then used this data to train a machine learning model that can accurately predict a player's impact on the game. We further passed the obtained results as one of the features to the predictor of transfer fees. The model can help clubs identify players who are undervalued and who could be sold for a profit. It can also help clubs avoid overpaying for players. We believe that our model can be a valuable tool for football clubs. It can help them make better decisions about player recruitment and transfers." }, "2402.00163.pdf": { "id": "2402.00163v1", "title": "Improving Object Detection Quality in Football Through Super-Resolution Techniques", "year": "2024", "authors": [ "Karolina Seweryn", "Gabriel Ch\u0119\u0107", "Szymon \u0141ukasik", "Anna Wr\u00f3blewska" ], "categories": "cs.CV, cs.AI", "pdf_url": "https://arxiv.org/pdf/2402.00163v1", "abstract": "This study explores the potential of super-resolution techniques in enhancing object detection accuracy in football. Given the sport's fast-paced nature and the critical importance of precise object (e.g. ball, player) tracking for both analysis and broadcasting, super-resolution could offer significant improvements. We investigate how advanced image processing through super-resolution impacts the accuracy and reliability of object detection algorithms in processing football match footage. Our methodology involved applying state-of-the-art super-resolution techniques to a diverse set of football match videos from SoccerNet, followed by object detection using Faster R-CNN. The performance of these algorithms, both with and without super-resolution enhancement, was rigorously evaluated in terms of detection accuracy. The results indicate a marked improvement in object detection accuracy when super-resolution preprocessing is applied. The improvement of object detection through the integration of super-resolution techniques yields significant benefits, especially for low-resolution scenarios, with a notable 12\\% increase in mean Average Precision (mAP) at an IoU (Intersection over Union) range of 0.50:0.95 for 320x240 size images when increasing the resolution fourfold using RLFN. As the dimensions increase, the magnitude of improvement becomes more subdued; however, a discernible improvement in the quality of detection is consistently evident. Additionally, we discuss the implications of these findings for real-time sports analytics, player tracking, and the overall viewing experience. The study contributes to the growing field of sports technology by demonstrating the practical benefits and limitations of integrating super-resolution techniques in football analytics and broadcasting." }, "2402.04898.pdf": { "id": "2402.04898v1", "title": "The Strain of Success: A Predictive Model for Injury Risk Mitigation and Team Success in Soccer", "year": "2024", "authors": [ "Gregory Everett", "Ryan Beal", "Tim Matthews", "Timothy J. Norman", "Sarvapali D. Ramchurn" ], "categories": "cs.AI, cs.LG", "pdf_url": "https://arxiv.org/pdf/2402.04898v1", "abstract": "In this paper, we present a novel sequential team selection model in soccer. Specifically, we model the stochastic process of player injury and unavailability using player-specific information learned from real-world soccer data. Monte-Carlo Tree Search is used to select teams for games that optimise long-term team performance across a soccer season by reasoning over player injury probability. We validate our approach compared to benchmark solutions for the 2018/19 English Premier League season. Our model achieves similar season expected points to the benchmark whilst reducing first-team injuries by ~13% and the money inefficiently spent on injured players by ~11% - demonstrating the potential to reduce costs and improve player welfare in real-world soccer teams." }, "2402.06815.pdf": { "id": "2402.06815v2", "title": "Estimating Player Performance in Different Contexts Using Fine-tuned Large Events Models", "year": "2024", "authors": [ "Tiago Mendes-Neves", "Lu\u00eds Meireles", "Jo\u00e3o Mendes-Moreira" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2402.06815v2", "abstract": "This paper introduces an innovative application of Large Event Models (LEMs), akin to Large Language Models, to the domain of soccer analytics. By learning the language of soccer - predicting variables for subsequent events rather than words - LEMs facilitate the simulation of matches and offer various applications, including player performance prediction across different team contexts. We focus on fine-tuning LEMs with the WyScout dataset for the 2017-2018 Premier League season to derive specific insights into player contributions and team strategies. Our methodology involves adapting these models to reflect the nuanced dynamics of soccer, enabling the evaluation of hypothetical transfers. Our findings confirm the effectiveness and limitations of LEMs in soccer analytics, highlighting the model's capability to forecast teams' expected standings and explore high-profile scenarios, such as the potential effects of transferring Cristiano Ronaldo or Lionel Messi to different teams in the Premier League. This analysis underscores the importance of context in evaluating player quality. While general metrics may suggest significant differences between players, contextual analyses reveal narrower gaps in performance within specific team frameworks." }, "2402.06820.pdf": { "id": "2402.06820v2", "title": "Forecasting Events in Soccer Matches Through Language", "year": "2024", "authors": [ "Tiago Mendes-Neves", "Lu\u00eds Meireles", "Jo\u00e3o Mendes-Moreira" ], "categories": "cs.LG, cs.CL", "pdf_url": "https://arxiv.org/pdf/2402.06820v2", "abstract": "This paper introduces an approach to predicting the next event in a soccer match, a challenge bearing remarkable similarities to the problem faced by Large Language Models (LLMs). Unlike other methods that severely limit event dynamics in soccer, often abstracting from many variables or relying on a mix of sequential models, our research proposes a novel technique inspired by the methodologies used in LLMs. These models predict a complete chain of variables that compose an event, significantly simplifying the construction of Large Event Models (LEMs) for soccer. Utilizing deep learning on the publicly available WyScout dataset, the proposed approach notably surpasses the performance of previous LEM proposals in critical areas, such as the prediction accuracy of the next event type. This paper highlights the utility of LEMs in various applications, including match prediction and analytics. Moreover, we show that LEMs provide a simulation backbone for users to build many analytics pipelines, an approach opposite to the current specialized single-purpose models. LEMs represent a pivotal advancement in soccer analytics, establishing a foundational framework for multifaceted analytics pipelines through a singular machine-learning model." }, "2402.09650.pdf": { "id": "2402.09650v1", "title": "Foul prediction with estimated poses from soccer broadcast video", "year": "2024", "authors": [ "Jiale Fang", "Calvin Yeung", "Keisuke Fujii" ], "categories": "cs.CV, cs.LG", "pdf_url": "https://arxiv.org/pdf/2402.09650v1", "abstract": "Recent advances in computer vision have made significant progress in tracking and pose estimation of sports players. However, there have been fewer studies on behavior prediction with pose estimation in sports, in particular, the prediction of soccer fouls is challenging because of the smaller image size of each player and of difficulty in the usage of e.g., the ball and pose information. In our research, we introduce an innovative deep learning approach for anticipating soccer fouls. This method integrates video data, bounding box positions, image details, and pose information by curating a novel soccer foul dataset. Our model utilizes a combination of convolutional and recurrent neural networks (CNNs and RNNs) to effectively merge information from these four modalities. The experimental results show that our full model outperformed the ablated models, and all of the RNN modules, bounding box position and image, and estimated pose were useful for the foul prediction. Our findings have important implications for a deeper understanding of foul play in soccer and provide a valuable reference for future research and practice in this area." }, "2402.17930.pdf": { "id": "2402.17930v1", "title": "Pragmatic Instruction Following and Goal Assistance via Cooperative Language-Guided Inverse Planning", "year": "2024", "authors": [ "Tan Zhi-Xuan", "Lance Ying", "Vikash Mansinghka", "Joshua B. Tenenbaum" ], "categories": "cs.AI, cs.CL, cs.LG", "pdf_url": "https://arxiv.org/pdf/2402.17930v1", "abstract": "People often give instructions whose meaning is ambiguous without further context, expecting that their actions or goals will disambiguate their intentions. How can we build assistive agents that follow such instructions in a flexible, context-sensitive manner? This paper introduces cooperative language-guided inverse plan search (CLIPS), a Bayesian agent architecture for pragmatic instruction following and goal assistance. Our agent assists a human by modeling them as a cooperative planner who communicates joint plans to the assistant, then performs multimodal Bayesian inference over the human's goal from actions and language, using large language models (LLMs) to evaluate the likelihood of an instruction given a hypothesized plan. Given this posterior, our assistant acts to minimize expected goal achievement cost, enabling it to pragmatically follow ambiguous instructions and provide effective assistance even when uncertain about the goal. We evaluate these capabilities in two cooperative planning domains (Doors, Keys & Gems and VirtualHome), finding that CLIPS significantly outperforms GPT-4V, LLM-based literal instruction following and unimodal inverse planning in both accuracy and helpfulness, while closely matching the inferences and assistive judgments provided by human raters." }, "2403.01112.pdf": { "id": "2403.01112v2", "title": "Efficient Episodic Memory Utilization of Cooperative Multi-Agent Reinforcement Learning", "year": "2024", "authors": [ "Hyungho Na", "Yunkyeong Seo", "Il-chul Moon" ], "categories": "cs.LG, cs.MA", "pdf_url": "https://arxiv.org/pdf/2403.01112v2", "abstract": "In cooperative multi-agent reinforcement learning (MARL), agents aim to achieve a common goal, such as defeating enemies or scoring a goal. Existing MARL algorithms are effective but still require significant learning time and often get trapped in local optima by complex tasks, subsequently failing to discover a goal-reaching policy. To address this, we introduce Efficient episodic Memory Utilization (EMU) for MARL, with two primary objectives: (a) accelerating reinforcement learning by leveraging semantically coherent memory from an episodic buffer and (b) selectively promoting desirable transitions to prevent local convergence. To achieve (a), EMU incorporates a trainable encoder/decoder structure alongside MARL, creating coherent memory embeddings that facilitate exploratory memory recall. To achieve (b), EMU introduces a novel reward structure called episodic incentive based on the desirability of states. This reward improves the TD target in Q-learning and acts as an additional incentive for desirable transitions. We provide theoretical support for the proposed incentive and demonstrate the effectiveness of EMU compared to conventional episodic control. The proposed method is evaluated in StarCraft II and Google Research Football, and empirical results indicate further performance improvement over state-of-the-art methods." }, "2403.02741.pdf": { "id": "2403.02741v2", "title": "State-Constrained Zero-Sum Differential Games with One-Sided Information", "year": "2024", "authors": [ "Mukesh Ghimire", "Lei Zhang", "Zhe Xu", "Yi Ren" ], "categories": "cs.GT, cs.LG", "pdf_url": "https://arxiv.org/pdf/2403.02741v2", "abstract": "We study zero-sum differential games with state constraints and one-sided information, where the informed player (Player 1) has a categorical payoff type unknown to the uninformed player (Player 2). The goal of Player 1 is to minimize his payoff without violating the constraints, while that of Player 2 is to violate the state constraints if possible, or to maximize the payoff otherwise. One example of the game is a man-to-man matchup in football. Without state constraints, Cardaliaguet (2007) showed that the value of such a game exists and is convex to the common belief of players. Our theoretical contribution is an extension of this result to games with state constraints and the derivation of the primal and dual subdynamic principles necessary for computing behavioral strategies. Different from existing works that are concerned about the scalability of no-regret learning in games with discrete dynamics, our study reveals the underlying structure of strategies for belief manipulation resulting from information asymmetry and state constraints. This structure will be necessary for scalable learning on games with continuous actions and long time windows. We use a simplified football game to demonstrate the utility of this work, where we reveal player positions and belief states in which the attacker should (or should not) play specific random deceptive moves to take advantage of information asymmetry, and compute how the defender should respond." }, "2403.03172.pdf": { "id": "2403.03172v1", "title": "Reaching Consensus in Cooperative Multi-Agent Reinforcement Learning with Goal Imagination", "year": "2024", "authors": [ "Liangzhou Wang", "Kaiwen Zhu", "Fengming Zhu", "Xinghu Yao", "Shujie Zhang", "Deheng Ye" ], "categories": "cs.AI, cs.LG", "pdf_url": "https://arxiv.org/pdf/2403.03172v1", "abstract": "Reaching consensus is key to multi-agent coordination. To accomplish a cooperative task, agents need to coherently select optimal joint actions to maximize the team reward. However, current cooperative multi-agent reinforcement learning (MARL) methods usually do not explicitly take consensus into consideration, which may cause miscoordination problem. In this paper, we propose a model-based consensus mechanism to explicitly coordinate multiple agents. The proposed Multi-agent Goal Imagination (MAGI) framework guides agents to reach consensus with an Imagined common goal. The common goal is an achievable state with high value, which is obtained by sampling from the distribution of future states. We directly model this distribution with a self-supervised generative model, thus alleviating the \"curse of dimensinality\" problem induced by multi-agent multi-step policy rollout commonly used in model-based methods. We show that such efficient consensus mechanism can guide all agents cooperatively reaching valuable future states. Results on Multi-agent Particle-Environments and Google Research Football environment demonstrate the superiority of MAGI in both sample efficiency and performance." }, "2403.07669.pdf": { "id": "2403.07669v1", "title": "Machine Learning for Soccer Match Result Prediction", "year": "2024", "authors": [ "Rory Bunker", "Calvin Yeung", "Keisuke Fujii" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2403.07669v1", "abstract": "Machine learning has become a common approach to predicting the outcomes of soccer matches, and the body of literature in this domain has grown substantially in the past decade and a half. This chapter discusses available datasets, the types of models and features, and ways of evaluating model performance in this application domain. The aim of this chapter is to give a broad overview of the current state and potential future developments in machine learning for soccer match results prediction, as a resource for those interested in conducting future studies in the area. Our main findings are that while gradient-boosted tree models such as CatBoost, applied to soccer-specific ratings such as pi-ratings, are currently the best-performing models on datasets containing only goals as the match features, there needs to be a more thorough comparison of the performance of deep learning models and Random Forest on a range of datasets with different types of features. Furthermore, new rating systems using both player- and team-level information and incorporating additional information from, e.g., spatiotemporal tracking and event data, could be investigated further. Finally, the interpretability of match result prediction models needs to be enhanced for them to be more useful for team management." }, "2403.08835.pdf": { "id": "2403.08835v1", "title": "Stacking-based deep neural network for player scouting in football 1", "year": "2024", "authors": [ "Simon Lacan" ], "categories": "cs.LG, cs.AI", "pdf_url": "https://arxiv.org/pdf/2403.08835v1", "abstract": "Datascouting is one of the most known data applications in professional sport, and specifically football. Its objective is to analyze huge database of players in order to detect high potentials that can be then individually considered by human scouts. In this paper, we propose a stacking-based deep learning model to detect high potential football players. Applied on open-source database, our model obtains significantly better results that classical statistical methods." }, "2403.11051.pdf": { "id": "2403.11051v1", "title": "Gender differences in online communication: A case study of Soccer", "year": "2024", "authors": [ "Mariana Macedo", "Akrati Saxena" ], "categories": "cs.SI", "pdf_url": "https://arxiv.org/pdf/2403.11051v1", "abstract": "Social media and digital platforms allow us to express our opinions freely and easily to a vast number of people. In this study, we examine whether there are gender-based differences in how communication happens via Twitter in regard to soccer. Soccer is one of the most popular sports, and therefore, on social media, it engages a diverse audience regardless of their technical knowledge. We collected Twitter data for three months (March-June) for English and Portuguese that contains 9.5 million Tweets related to soccer, and only 18.38% tweets were identified as belonging to women, highlighting a possible gender gap already in the number of people who participated actively in this topic. We then conduct a fine-grained text-level and network-level analysis to identify the gender differences that might exist while communicating on Twitter. Our results show that women express their emotions more intensely than men, regardless of the differences in volume. The network generated from Portuguese has lower homophily than English. However, this difference in homophily does not impact how females express their emotions and sentiments, suggesting that these aspects are inherent norms or characteristics of genders. Our study unveils more gaps through qualitative and quantitative analyses, highlighting the importance of examining and reporting gender gaps in online communication to create a more inclusive space where people can openly share their opinions." }, "2403.12977.pdf": { "id": "2403.12977v3", "title": "SportsNGEN: Sustained Generation of Realistic Multi-player Sports Gameplay", "year": "2024", "authors": [ "Lachlan Thorpe", "Lewis Bawden", "Karanjot Vendal", "John Bronskill", "Richard E. Turner" ], "categories": "cs.CV, cs.LG, eess.IV, stat.AP", "pdf_url": "https://arxiv.org/pdf/2403.12977v3", "abstract": "We present a transformer decoder based sports simulation engine, SportsNGEN, trained on sports player and ball tracking sequences, that is capable of generating sustained gameplay and accurately mimicking the decision making of real players. By training on a large database of professional tennis tracking data, we demonstrate that simulations produced by SportsNGEN can be used to predict the outcomes of rallies, determine the best shot choices at any point, and evaluate counterfactual or what if scenarios to inform coaching decisions and elevate broadcast coverage. By combining the generated simulations with a shot classifier and logic to start and end rallies, the system is capable of simulating an entire tennis match. We evaluate SportsNGEN by comparing statistics of the simulations with those of real matches between the same players. We show that the model output sampling parameters are crucial to simulation realism and that SportsNGEN is probabilistically well-calibrated to real data. In addition, a generic version of SportsNGEN can be customized to a specific player by fine-tuning on the subset of match data that includes that player. Finally, we show qualitative results indicating the same approach works for football." }, "2403.14300.pdf": { "id": "2403.14300v1", "title": "DexDribbler: Learning Dexterous Soccer Manipulation via Dynamic Supervision", "year": "2024", "authors": [ "Yutong Hu", "Kehan Wen", "Fisher Yu" ], "categories": "cs.RO, cs.AI", "pdf_url": "https://arxiv.org/pdf/2403.14300v1", "abstract": "Learning dexterous locomotion policy for legged robots is becoming increasingly popular due to its ability to handle diverse terrains and resemble intelligent behaviors. However, joint manipulation of moving objects and locomotion with legs, such as playing soccer, receive scant attention in the learning community, although it is natural for humans and smart animals. A key challenge to solve this multitask problem is to infer the objectives of locomotion from the states and targets of the manipulated objects. The implicit relation between the object states and robot locomotion can be hard to capture directly from the training experience. We propose adding a feedback control block to compute the necessary body-level movement accurately and using the outputs as dynamic joint-level locomotion supervision explicitly. We further utilize an improved ball dynamic model, an extended context-aided estimator, and a comprehensive ball observer to facilitate transferring policy learned in simulation to the real world. We observe that our learning scheme can not only make the policy network converge faster but also enable soccer robots to perform sophisticated maneuvers like sharp cuts and turns on flat surfaces, a capability that was lacking in previous methods. Video and code are available at https://github.com/SysCV/soccer-player" }, "2403.15200.pdf": { "id": "2403.15200v1", "title": "Teamwork and Spillover Effects in Performance Evaluations", "year": "2024", "authors": [ "Enzo Brox", "Michael Lechner" ], "categories": "econ.GN", "pdf_url": "https://arxiv.org/pdf/2403.15200v1", "abstract": "This article shows how coworker performance affects individual performance evaluation in a teamwork setting at the workplace. We use high-quality data on football matches to measure an important component of individual performance, shooting performance, isolated from collaborative effects. Employing causal machine learning methods, we address the assortative matching of workers and estimate both average and heterogeneous effects. There is substantial evidence for spillover effects in performance evaluations. Coworker shooting performance, meaningfully impacts both, manager decisions and third-party expert evaluations of individual performance. Our results underscore the significant role coworkers play in shaping career advancements and highlight a complementary channel, to productivity gains and learning effects, how coworkers impact career advancement. We characterize the groups of workers that are most and least affected by spillover effects and show that spillover effects are reference point dependent. While positive deviations from a reference point create positive spillover effects, negative deviations are not harmful for coworkers." }, "2403.16282.pdf": { "id": "2403.16282v1", "title": "The Evolution of Football Betting- A Machine Learning Approach to Match Outcome Forecasting and Bookmaker Odds Estimation", "year": "2024", "authors": [ "Purnachandra Mandadapu" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2403.16282v1", "abstract": "This paper explores the significant history of professional football and the betting industry, tracing its evolution from clandestine beginnings to a lucrative multi-million-pound enterprise. Initiated by the legalization of gambling in 1960 and complemented by advancements in football data gathering pioneered by Thorold Charles Reep, the symbiotic relationship between these sectors has propelled rapid growth and innovation. Over the past six decades, both industries have undergone radical transformations, with data collection methods evolving from rudimentary notetaking to sophisticated technologies such as high-definition cameras and Artificial Intelligence (AI)-driven analytics. Therefore, the primary aim of this study is to utilize Machine Learning (ML) algorithms to forecast premier league football match outcomes. By analyzing historical data and investigating the significance of various features, the study seeks to identify the most effective predictive models and discern key factors influencing match results. Additionally, the study aims to utilize these forecasting to inform the establishment of bookmaker odds, providing insights into the impact of different variables on match outcomes. By highlighting the potential for informed decision-making in sports forecasting and betting, this study opens up new avenues for research and practical applications in the domain of sports analytics." }, "2404.01891.pdf": { "id": "2404.01891v1", "title": "ASTRA: An Action Spotting TRAnsformer for Soccer Videos", "year": "2024", "authors": [ "Artur Xarles", "Sergio Escalera", "Thomas B. Moeslund", "Albert Clap\u00e9s" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2404.01891v1", "abstract": "In this paper, we introduce ASTRA, a Transformer-based model designed for the task of Action Spotting in soccer matches. ASTRA addresses several challenges inherent in the task and dataset, including the requirement for precise action localization, the presence of a long-tail data distribution, non-visibility in certain actions, and inherent label noise. To do so, ASTRA incorporates (a) a Transformer encoder-decoder architecture to achieve the desired output temporal resolution and to produce precise predictions, (b) a balanced mixup strategy to handle the long-tail distribution of the data, (c) an uncertainty-aware displacement head to capture the label variability, and (d) input audio signal to enhance detection of non-visible actions. Results demonstrate the effectiveness of ASTRA, achieving a tight Average-mAP of 66.82 on the test set. Moreover, in the SoccerNet 2023 Action Spotting challenge, we secure the 3rd position with an Average-mAP of 70.21 on the challenge set." }, "2404.03101.pdf": { "id": "2404.03101v1", "title": "MARL-LNS: Cooperative Multi-agent Reinforcement Learning via Large Neighborhoods Search", "year": "2024", "authors": [ "Weizhe Chen", "Sven Koenig", "Bistra Dilkina" ], "categories": "cs.MA, cs.LG", "pdf_url": "https://arxiv.org/pdf/2404.03101v1", "abstract": "Cooperative multi-agent reinforcement learning (MARL) has been an increasingly important research topic in the last half-decade because of its great potential for real-world applications. Because of the curse of dimensionality, the popular \"centralized training decentralized execution\" framework requires a long time in training, yet still cannot converge efficiently. In this paper, we propose a general training framework, MARL-LNS, to algorithmically address these issues by training on alternating subsets of agents using existing deep MARL algorithms as low-level trainers, while not involving any additional parameters to be trained. Based on this framework, we provide three algorithm variants based on the framework: random large neighborhood search (RLNS), batch large neighborhood search (BLNS), and adaptive large neighborhood search (ALNS), which alternate the subsets of agents differently. We test our algorithms on both the StarCraft Multi-Agent Challenge and Google Research Football, showing that our algorithms can automatically reduce at least 10% of training time while reaching the same final skill level as the original algorithm." }, "2404.03869.pdf": { "id": "2404.03869v2", "title": "Heterogeneous Multi-Agent Reinforcement Learning for Zero-Shot Scalable Collaboration", "year": "2024", "authors": [ "Xudong Guo", "Daming Shi", "Junjie Yu", "Wenhui Fan" ], "categories": "cs.LG, cs.AI, cs.MA, cs.RO, eess.SY", "pdf_url": "https://arxiv.org/pdf/2404.03869v2", "abstract": "The emergence of multi-agent reinforcement learning (MARL) is significantly transforming various fields like autonomous vehicle networks. However, real-world multi-agent systems typically contain multiple roles, and the scale of these systems dynamically fluctuates. Consequently, in order to achieve zero-shot scalable collaboration, it is essential that strategies for different roles can be updated flexibly according to the scales, which is still a challenge for current MARL frameworks. To address this, we propose a novel MARL framework named Scalable and Heterogeneous Proximal Policy Optimization (SHPPO), integrating heterogeneity into parameter-shared PPO-based MARL networks. We first leverage a latent network to learn strategy patterns for each agent adaptively. Second, we introduce a heterogeneous layer to be inserted into decision-making networks, whose parameters are specifically generated by the learned latent variables. Our approach is scalable as all the parameters are shared except for the heterogeneous layer, and gains both inter-individual and temporal heterogeneity, allowing SHPPO to adapt effectively to varying scales. SHPPO exhibits superior performance in classic MARL environments like Starcraft Multi-Agent Challenge (SMAC) and Google Research Football (GRF), showcasing enhanced zero-shot scalability, and offering insights into the learned latent variables' impact on team performance by visualization." }, "2405.02412.pdf": { "id": "2405.02412v1", "title": "Deep Learning and Transfer Learning Architectures for English Premier League Player Performance Forecasting", "year": "2024", "authors": [ "Daniel Frees", "Pranav Ravella", "Charlie Zhang" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2405.02412v1", "abstract": "This paper presents a groundbreaking model for forecasting English Premier League (EPL) player performance using convolutional neural networks (CNNs). We evaluate Ridge regression, LightGBM and CNNs on the task of predicting upcoming player FPL score based on historical FPL data over the previous weeks. Our baseline models, Ridge regression and LightGBM, achieve solid performance and emphasize the importance of recent FPL points, influence, creativity, threat, and playtime in predicting EPL player performances. Our optimal CNN architecture achieves better performance with fewer input features and even outperforms the best previous EPL player performance forecasting models in the literature. The optimal CNN architecture also achieves very strong Spearman correlation with player rankings, indicating its strong implications for supporting the development of FPL artificial intelligence (AI) Agents and providing analysis for FPL managers. We additionally perform transfer learning experiments on soccer news data collected from The Guardian, for the same task of predicting upcoming player score, but do not identify a strong predictive signal in natural language news texts, achieving worse performance compared to both the CNN and baseline models. Overall, our CNN-based approach marks a significant advancement in EPL player performance forecasting and lays the foundation for transfer learning to other EPL prediction tasks such as win-loss odds for sports betting and the development of cutting-edge FPL AI Agents." }, "2405.02425.pdf": { "id": "2405.02425v1", "title": "Learning Robot Soccer from Egocentric Vision with Deep Reinforcement Learning", "year": "2024", "authors": [ "Dhruva Tirumala", "Markus Wulfmeier", "Ben Moran", "Sandy Huang", "Jan Humplik", "Guy Lever" ], "categories": "cs.RO, cs.AI", "pdf_url": "https://arxiv.org/pdf/2405.02425v1", "abstract": "We apply multi-agent deep reinforcement learning (RL) to train end-to-end robot soccer policies with fully onboard computation and sensing via egocentric RGB vision. This setting reflects many challenges of real-world robotics, including active perception, agile full-body control, and long-horizon planning in a dynamic, partially-observable, multi-agent domain. We rely on large-scale, simulation-based data generation to obtain complex behaviors from egocentric vision which can be successfully transferred to physical robots using low-cost sensors. To achieve adequate visual realism, our simulation combines rigid-body physics with learned, realistic rendering via multiple Neural Radiance Fields (NeRFs). We combine teacher-based multi-agent RL and cross-experiment data reuse to enable the discovery of sophisticated soccer strategies. We analyze active-perception behaviors including object tracking and ball seeking that emerge when simply optimizing perception-agnostic soccer play. The agents display equivalent levels of performance and agility as policies with access to privileged, ground-truth state. To our knowledge, this paper constitutes a first demonstration of end-to-end training for multi-agent robot soccer, mapping raw pixel observations to joint-level actions, that can be deployed in the real world. Videos of the game-play and analyses can be seen on our website https://sites.google.com/view/vision-soccer ." }, "2405.07354.pdf": { "id": "2405.07354v1", "title": "SoccerNet-Echoes: A Soccer Game Audio Commentary Dataset", "year": "2024", "authors": [ "Sushant Gautam", "Mehdi Houshmand Sarkhoosh", "Jan Held", "Cise Midoglu", "Anthony Cioppa", "Silvio Giancola" ], "categories": "cs.SD, cs.IR, cs.LG, cs.MM, eess.AS", "pdf_url": "https://arxiv.org/pdf/2405.07354v1", "abstract": "The application of Automatic Speech Recognition (ASR) technology in soccer offers numerous opportunities for sports analytics. Specifically, extracting audio commentaries with ASR provides valuable insights into the events of the game, and opens the door to several downstream applications such as automatic highlight generation. This paper presents SoccerNet-Echoes, an augmentation of the SoccerNet dataset with automatically generated transcriptions of audio commentaries from soccer game broadcasts, enhancing video content with rich layers of textual information derived from the game audio using ASR. These textual commentaries, generated using the Whisper model and translated with Google Translate, extend the usefulness of the SoccerNet dataset in diverse applications such as enhanced action spotting, automatic caption generation, and game summarization. By incorporating textual data alongside visual and auditory content, SoccerNet-Echoes aims to serve as a comprehensive resource for the development of algorithms specialized in capturing the dynamics of soccer games. We detail the methods involved in the curation of this dataset and the integration of ASR. We also highlight the implications of a multimodal approach in sports analytics, and how the enriched dataset can support diverse applications, thus broadening the scope of research and development in the field of sports analytics." }, "2405.10247.pdf": { "id": "2405.10247v3", "title": "Alternative ranking measures to predict international football results", "year": "2024", "authors": [ "Roberto Macr\u00ec Demartino", "Leonardo Egidi", "Nicola Torelli" ], "categories": "stat.AP", "pdf_url": "https://arxiv.org/pdf/2405.10247v3", "abstract": "Over the last few years, there has been a growing interest in the prediction and modelling of competitive sports outcomes, with particular emphasis placed on this area by the Bayesian statistics and machine learning communities. In this paper, we have carried out a comparative evaluation of statistical and machine learning models to assess their predictive performance for the 2022 FIFA World Cup and for the 2023 CAF Africa Cup of Nations by evaluating alternative summaries of past performances related to the involved teams. More specifically, we consider the Bayesian Bradley-Terry-Davidson model, which is a widely used statistical framework for ranking items based on paired comparisons that have been applied successfully in various domains, including football. The analysis was performed including in some canonical goal-based models both the Bradley-Terry-Davidson derived ranking and the widely recognized Coca-Cola FIFA ranking commonly adopted by football fans and amateurs." }, "2405.11746.pdf": { "id": "2405.11746v1", "title": "Configurable Mirror Descent: Towards a Unification of Decision Making", "year": "2024", "authors": [ "Pengdeng Li", "Shuxin Li", "Chang Yang", "Xinrun Wang", "Shuyue Hu", "Xiao Huang" ], "categories": "cs.AI, cs.GT, cs.LG, cs.MA", "pdf_url": "https://arxiv.org/pdf/2405.11746v1", "abstract": "Decision-making problems, categorized as single-agent, e.g., Atari, cooperative multi-agent, e.g., Hanabi, competitive multi-agent, e.g., Hold'em poker, and mixed cooperative and competitive, e.g., football, are ubiquitous in the real world. Various methods are proposed to address the specific decision-making problems. Despite the successes in specific categories, these methods typically evolve independently and cannot generalize to other categories. Therefore, a fundamental question for decision-making is: \\emph{Can we develop \\textbf{a single algorithm} to tackle \\textbf{ALL} categories of decision-making problems?} There are several main challenges to address this question: i) different decision-making categories involve different numbers of agents and different relationships between agents, ii) different categories have different solution concepts and evaluation measures, and iii) there lacks a comprehensive benchmark covering all the categories. This work presents a preliminary attempt to address the question with three main contributions. i) We propose the generalized mirror descent (GMD), a generalization of MD variants, which considers multiple historical policies and works with a broader class of Bregman divergences. ii) We propose the configurable mirror descent (CMD) where a meta-controller is introduced to dynamically adjust the hyper-parameters in GMD conditional on the evaluation measures. iii) We construct the \\textsc{GameBench} with 15 academic-friendly games across different decision-making categories. Extensive experiments demonstrate that CMD achieves empirically competitive or better outcomes compared to baselines while providing the capability of exploring diverse dimensions of decision making." }, "2405.12070.pdf": { "id": "2405.12070v1", "title": "AutoSoccerPose: Automated 3D posture Analysis of Soccer Shot Movements", "year": "2024", "authors": [ "Calvin Yeung", "Kenjiro Ide", "Keisuke Fujii" ], "categories": "cs.CV, cs.AI", "pdf_url": "https://arxiv.org/pdf/2405.12070v1", "abstract": "Image understanding is a foundational task in computer vision, with recent applications emerging in soccer posture analysis. However, existing publicly available datasets lack comprehensive information, notably in the form of posture sequences and 2D pose annotations. Moreover, current analysis models often rely on interpretable linear models (e.g., PCA and regression), limiting their capacity to capture non-linear spatiotemporal relationships in complex and diverse scenarios. To address these gaps, we introduce the 3D Shot Posture (3DSP) dataset in soccer broadcast videos, which represents the most extensive sports image dataset with 2D pose annotations to our knowledge. Additionally, we present the 3DSP-GRAE (Graph Recurrent AutoEncoder) model, a non-linear approach for embedding pose sequences. Furthermore, we propose AutoSoccerPose, a pipeline aimed at semi-automating 2D and 3D pose estimation and posture analysis. While achieving full automation proved challenging, we provide a foundational baseline, extending its utility beyond the scope of annotated data. We validate AutoSoccerPose on SoccerNet and 3DSP datasets, and present posture analysis results based on 3DSP. The dataset, code, and models are available at: https://github.com/calvinyeungck/3D-Shot-Posture-Dataset." }, "2405.12588.pdf": { "id": "2405.12588v1", "title": "An analysis of factors impacting team strengths in the Australian Football League using time-variant Bradley-Terry models", "year": "2024", "authors": [ "Carlos Rafael Gonzalez Soffner", "Manuele Leonelli" ], "categories": "stat.AP", "pdf_url": "https://arxiv.org/pdf/2405.12588v1", "abstract": "Australian Rules Football is a field invasion game where two teams attempt to score the highest points to win. Complex machine learning algorithms have been developed to predict match outcomes post-game, but their lack of interpretability hampers an understanding of the factors that affect a team's performance. Using data from the male competition of the Australian Football League, seasons 2015 to 2023, we estimate team strengths and the factors impacting them by fitting flexible Bradley-Terry models. We successfully identify teams significantly stronger or weaker than the average, with stronger teams placing higher in the previous seasons' ladder and leading the activity in the Forward 50 zone, goal shots and scoring over their opponents. Playing at home is confirmed to create an advantage regardless of team strengths. The ability of the model to predict game results in advance is tested, with models accounting for team-specific, time-variant features predicting up to 71.5% of outcomes. Therefore, our approach can provide an interpretable understanding of team strengths and competitive game predictions, making it optimal for data-driven strategies and training." }, "2405.12628.pdf": { "id": "2405.12628v1", "title": "Play Everywhere: A Temporal Logic based Game Environment Independent Approach for Playing Soccer with Robots", "year": "2024", "authors": [ "Vincenzo Suriani", "Emanuele Musumeci", "Daniele Nardi", "Domenico Daniele Bloisi" ], "categories": "cs.RO, cs.AI", "pdf_url": "https://arxiv.org/pdf/2405.12628v1", "abstract": "Robots playing soccer often rely on hard-coded behaviors that struggle to generalize when the game environment change. In this paper, we propose a temporal logic based approach that allows robots' behaviors and goals to adapt to the semantics of the environment. In particular, we present a hierarchical representation of soccer in which the robot selects the level of operation based on the perceived semantic characteristics of the environment, thus modifying dynamically the set of rules and goals to apply. The proposed approach enables the robot to operate in unstructured environments, just as it happens when humans go from soccer played on an official field to soccer played on a street. Three different use cases set in different scenarios are presented to demonstrate the effectiveness of the proposed approach." }, "2405.17680.pdf": { "id": "2405.17680v2", "title": "Sports-Traj: A Unified Trajectory Generation Model for Multi-Agent Movement in Sports", "year": "2024", "authors": [ "Yi Xu", "Yun Fu" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2405.17680v2", "abstract": "Understanding multi-agent movement is critical across various fields. The conventional approaches typically focus on separate tasks such as trajectory prediction, imputation, or spatial-temporal recovery. Considering the unique formulation and constraint of each task, most existing methods are tailored for only one, limiting the ability to handle multiple tasks simultaneously, which is a common requirement in real-world scenarios. Another limitation is that widely used public datasets mainly focus on pedestrian movements with casual, loosely connected patterns, where interactions between individuals are not always present, especially at a long distance, making them less representative of more structured environments. To overcome these limitations, we propose a Unified Trajectory Generation model, UniTraj, that processes arbitrary trajectories as masked inputs, adaptable to diverse scenarios in the domain of sports games. Specifically, we introduce a Ghost Spatial Masking (GSM) module, embedded within a Transformer encoder, for spatial feature extraction. We further extend recent State Space Models (SSMs), known as the Mamba model, into a Bidirectional Temporal Mamba (BTM) to better capture temporal dependencies. Additionally, we incorporate a Bidirectional Temporal Scaled (BTS) module to thoroughly scan trajectories while preserving temporal missing relationships. Furthermore, we curate and benchmark three practical sports datasets, Basketball-U, Football-U, and Soccer-U, for evaluation. Extensive experiments demonstrate the superior performance of our model. We hope that our work can advance the understanding of human movement in real-world applications, particularly in sports. Our datasets, code, and model weights are available here https://github.com/colorfulfuture/UniTraj-pytorch." }, "2405.18110.pdf": { "id": "2405.18110v1", "title": "Individual Contributions as Intrinsic Exploration Scaffolds for Multi-agent Reinforcement Learning", "year": "2024", "authors": [ "Xinran Li", "Zifan Liu", "Shibo Chen", "Jun Zhang" ], "categories": "cs.LG, cs.AI, cs.MA", "pdf_url": "https://arxiv.org/pdf/2405.18110v1", "abstract": "In multi-agent reinforcement learning (MARL), effective exploration is critical, especially in sparse reward environments. Although introducing global intrinsic rewards can foster exploration in such settings, it often complicates credit assignment among agents. To address this difficulty, we propose Individual Contributions as intrinsic Exploration Scaffolds (ICES), a novel approach to motivate exploration by assessing each agent's contribution from a global view. In particular, ICES constructs exploration scaffolds with Bayesian surprise, leveraging global transition information during centralized training. These scaffolds, used only in training, help to guide individual agents towards actions that significantly impact the global latent state transitions. Additionally, ICES separates exploration policies from exploitation policies, enabling the former to utilize privileged global information during training. Extensive experiments on cooperative benchmark tasks with sparse rewards, including Google Research Football (GRF) and StarCraft Multi-agent Challenge (SMAC), demonstrate that ICES exhibits superior exploration capabilities compared with baselines. The code is publicly available at https://github.com/LXXXXR/ICES." }, "2406.00814.pdf": { "id": "2406.00814v1", "title": "Expected Possession Value of Control and Duel Actions for Soccer Player's Skills Estimation", "year": "2024", "authors": [ "Andrei Shelopugin" ], "categories": "cs.LG, cs.AI", "pdf_url": "https://arxiv.org/pdf/2406.00814v1", "abstract": "Estimation of football players' skills is one of the key tasks in sports analytics. This paper introduces multiple extensions to a widely used model, expected possession value (EPV), to address some key challenges such as selection problem. First, we assign greater weights to events occurring immediately prior to the shot rather than those preceding them (decay effect). Second, our model incorporates possession risk more accurately by considering the decay effect and effective playing time. Third, we integrate the assessment of individual player ability to win aerial and ground duels. Using the extended EPV model, we predict this metric for various football players for the upcoming season, particularly taking into account the strength of their opponents." }, "2406.01033.pdf": { "id": "2406.01033v1", "title": "Generalized Jersey Number Recognition Using Multi-task Learning With Orientation-guided Weight Refinement", "year": "2024", "authors": [ "Yung-Hui Lin", "Yu-Wen Chang", "Huang-Chia Shih", "Takahiro Ogawa" ], "categories": "cs.CV, cs.LG, cs.MM", "pdf_url": "https://arxiv.org/pdf/2406.01033v1", "abstract": "Jersey number recognition (JNR) has always been an important task in sports analytics. Improving recognition accuracy remains an ongoing challenge because images are subject to blurring, occlusion, deformity, and low resolution. Recent research has addressed these problems using number localization and optical character recognition. Some approaches apply player identification schemes to image sequences, ignoring the impact of human body rotation angles on jersey digit identification. Accurately predicting the number of jersey digits by using a multi-task scheme to recognize each individual digit enables more robust results. Based on the above considerations, this paper proposes a multi-task learning method called the angle-digit refine scheme (ADRS), which combines human body orientation angles and digit number clues to recognize athletic jersey numbers. Based on our experimental results, our approach increases inference information, significantly improving prediction accuracy. Compared to state-of-the-art methods, which can only handle a single type of sport, the proposed method produces a more diverse and practical JNR application. The incorporation of diverse types of team sports such as soccer, football, basketball, volleyball, and baseball into our dataset contributes greatly to generalized JNR in sports analytics. Our accuracy achieves 64.07% on Top-1 and 89.97% on Top-2, with corresponding F1 scores of 67.46% and 90.64%, respectively." }, "2406.01273.pdf": { "id": "2406.01273v2", "title": "SoccerRAG: Multimodal Soccer Information Retrieval via Natural Queries", "year": "2024", "authors": [ "Aleksander Theo Strand", "Sushant Gautam", "Cise Midoglu", "P\u00e5l Halvorsen" ], "categories": "cs.IR, cs.MM", "pdf_url": "https://arxiv.org/pdf/2406.01273v2", "abstract": "The rapid evolution of digital sports media necessitates sophisticated information retrieval systems that can efficiently parse extensive multimodal datasets. This paper introduces SoccerRAG, an innovative framework designed to harness the power of Retrieval Augmented Generation (RAG) and Large Language Models (LLMs) to extract soccer-related information through natural language queries. By leveraging a multimodal dataset, SoccerRAG supports dynamic querying and automatic data validation, enhancing user interaction and accessibility to sports archives. Our evaluations indicate that SoccerRAG effectively handles complex queries, offering significant improvements over traditional retrieval systems in terms of accuracy and user engagement. The results underscore the potential of using RAG and LLMs in sports analytics, paving the way for future advancements in the accessibility and real-time processing of sports data." }, "2406.01280.pdf": { "id": "2406.01280v2", "title": "Demo: Soccer Information Retrieval via Natural Queries using SoccerRAG", "year": "2024", "authors": [ "Aleksander Theo Strand", "Sushant Gautam", "Cise Midoglu", "P\u00e5l Halvorsen" ], "categories": "cs.IR, cs.MM", "pdf_url": "https://arxiv.org/pdf/2406.01280v2", "abstract": "The rapid evolution of digital sports media necessitates sophisticated information retrieval systems that can efficiently parse extensive multimodal datasets. This paper demonstrates SoccerRAG, an innovative framework designed to harness the power of Retrieval Augmented Generation (RAG) and Large Language Models (LLMs) to extract soccer-related information through natural language queries. By leveraging a multimodal dataset, SoccerRAG supports dynamic querying and automatic data validation, enhancing user interaction and accessibility to sports archives. We present a novel interactive user interface (UI) based on the Chainlit framework which wraps around the core functionality, and enable users to interact with the SoccerRAG framework in a chatbot-like visual manner." }, "2406.05621.pdf": { "id": "2406.05621v1", "title": "Cross Language Soccer Framework: An Open Source Framework for the RoboCup 2D Soccer Simulation", "year": "2024", "authors": [ "Nader Zare", "Aref Sayareh", "Alireza Sadraii", "Arad Firouzkouhi", "Amilcar Soares" ], "categories": "cs.RO, cs.AI", "pdf_url": "https://arxiv.org/pdf/2406.05621v1", "abstract": "RoboCup Soccer Simulation 2D (SS2D) research is hampered by the complexity of existing Cpp-based codes like Helios, Cyrus, and Gliders, which also suffer from limited integration with modern machine learning frameworks. This development paper introduces a transformative solution a gRPC-based, language-agnostic framework that seamlessly integrates with the high-performance Helios base code. This approach not only facilitates the use of diverse programming languages including CSharp, JavaScript, and Python but also maintains the computational efficiency critical for real time decision making in SS2D. By breaking down language barriers, our framework significantly enhances collaborative potential and flexibility, empowering researchers to innovate without the overhead of mastering or developing extensive base codes. We invite the global research community to leverage and contribute to the Cross Language Soccer (CLS) framework, which is openly available under the MIT License, to drive forward the capabilities of multi-agent systems in soccer simulations." }, "2406.05623.pdf": { "id": "2406.05623v1", "title": "Observation Denoising in CYRUS Soccer Simulation 2D Team For RoboCup 2024", "year": "2024", "authors": [ "Nader Zare", "Aref Sayareh", "Sadra Khanjari", "Arad Firouzkouhi" ], "categories": "cs.RO, cs.AI", "pdf_url": "https://arxiv.org/pdf/2406.05623v1", "abstract": "In the Soccer Simulation 2D environment, accurate observation is crucial for effective decision making. However, challenges such as partial observation and noisy data can hinder performance. To address these issues, we propose a denoising algorithm that leverages predictive modeling and intersection analysis to enhance the accuracy of observations. Our approach aims to mitigate the impact of noise and partial data, leading to improved gameplay performance. This paper presents the framework, implementation, and preliminary results of our algorithm, demonstrating its potential in refining observations in Soccer Simulation 2D. Cyrus 2D Team is using a combination of Helios, Gliders, and Cyrus base codes." }, "2406.08749.pdf": { "id": "2406.08749v2", "title": "Mathematical models for off-ball scoring prediction in basketball", "year": "2024", "authors": [ "Rikako Kono", "Keisuke Fujii" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2406.08749v2", "abstract": "In professional basketball, the accurate prediction of scoring opportunities based on strategic decision-making is crucial for spatial and player evaluations. However, traditional models often face challenges in accounting for the complexities of off-ball movements, which are essential for comprehensive performance evaluations. In this study, we propose two mathematical models to predict off-ball scoring opportunities in basketball, considering pass-to-score and dribble-to-score sequences: the Ball Movement for Off-ball Scoring (BMOS) and the Ball Intercept and Movement for Off-ball Scoring (BIMOS) models. The BMOS model adapts principles from the Off-Ball Scoring Opportunities (OBSO) model, originally designed for soccer, to basketball, whereas the BIMOS model also incorporates the likelihood of interception during ball movements. We evaluated these models using player tracking data from 630 NBA games in the 2015-2016 regular season, demonstrating that the BIMOS model outperforms the BMOS model in terms of team scoring prediction accuracy, while also highlighting its potential for further development. Overall, the BIMOS model provides valuable insights for tactical analysis and player evaluation in basketball." }, "2406.15659.pdf": { "id": "2406.15659v1", "title": "Contextual Sprint Classification in Soccer Based on Deep Learning", "year": "2024", "authors": [ "Hyunsung Kim", "Gun-Hee Joe", "Jinsung Yoon", "Sang-Ki Ko" ], "categories": "cs.LG, cs.MA", "pdf_url": "https://arxiv.org/pdf/2406.15659v1", "abstract": "The analysis of high-intensity runs (or sprints) in soccer has long been a topic of interest for sports science researchers and practitioners. In particular, recent studies suggested contextualizing sprints based on their tactical purposes to better understand the physical-tactical requirements of modern match-play. However, they have a limitation in scalability, as human experts have to manually classify hundreds of sprints for every match. To address this challenge, this paper proposes a deep learning framework for automatically classifying sprints in soccer into contextual categories. The proposed model covers the permutation-invariant and sequential nature of multi-agent trajectories in soccer by deploying Set Transformers and a bidirectional GRU. We train the model with category labels made through the collaboration of human annotators and a rule-based classifier. Experimental results show that our model classifies sprints in the test dataset into 15 categories with the accuracy of 77.65%, implying the potential of the proposed framework for facilitating the integrated analysis of soccer sprints at scale." }, "2406.16171.pdf": { "id": "2406.16171v5", "title": "Exploring the Difficulty of Estimating Win Probability: A Simulation Study", "year": "2024", "authors": [ "Ryan S. Brill", "Ronald Yurko", "Abraham J. Wyner" ], "categories": "stat.ME, stat.AP", "pdf_url": "https://arxiv.org/pdf/2406.16171v5", "abstract": "Estimating win probability is one of the classic modeling tasks of sports analytics. Many widely used win probability estimators use machine learning to fit the relationship between a binary win/loss outcome variable and certain game-state variables. To illustrate just how difficult it is to accurately fit such a model from noisy and highly correlated observational data, in this paper we conduct a simulation study. We create a simplified random walk version of football in which true win probability at each game-state is known, and we see how well a model recovers it. We find that the dependence structure of observational play-by-play data substantially inflates the bias and variance of estimators and lowers the effective sample size. Further, to achieve approximately valid marginal coverage, win probability confidence intervals need to be substantially wide. Concisely, these are high variance estimators subject to substantial uncertainty. Our findings are not unique to the particular application of estimating win probability; they are broadly applicable across sports analytics, as myriad other sports datasets are clustered into groups of observations that share the same outcome." }, "2406.18285.pdf": { "id": "2406.18285v1", "title": "LLCoach: Generating Robot Soccer Plans using Multi-Role Large Language Models", "year": "2024", "authors": [ "Michele Brienza", "Emanuele Musumeci", "Vincenzo Suriani", "Daniele Affinita", "Andrea Pennisi", "Daniele Nardi" ], "categories": "cs.RO", "pdf_url": "https://arxiv.org/pdf/2406.18285v1", "abstract": "The deployment of robots into human scenarios necessitates advanced planning strategies, particularly when we ask robots to operate in dynamic, unstructured environments. RoboCup offers the chance to deploy robots in one of those scenarios, a human-shaped game represented by a soccer match. In such scenarios, robots must operate using predefined behaviors that can fail in unpredictable conditions. This paper introduces a novel application of Large Language Models (LLMs) to address the challenge of generating actionable plans in such settings, specifically within the context of the RoboCup Standard Platform League (SPL) competitions where robots are required to autonomously execute soccer strategies that emerge from the interactions of individual agents. In particular, we propose a multi-role approach leveraging the capabilities of LLMs to generate and refine plans for a robotic soccer team. The potential of the proposed method is demonstrated through an experimental evaluation,carried out simulating multiple matches where robots with AI-generated plans play against robots running human-built code." }, "2406.18530.pdf": { "id": "2406.18530v2", "title": "MatchTime: Towards Automatic Soccer Game Commentary Generation", "year": "2024", "authors": [ "Jiayuan Rao", "Haoning Wu", "Chang Liu", "Yanfeng Wang", "Weidi Xie" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2406.18530v2", "abstract": "Soccer is a globally popular sport with a vast audience, in this paper, we consider constructing an automatic soccer game commentary model to improve the audiences' viewing experience. In general, we make the following contributions: First, observing the prevalent video-text misalignment in existing datasets, we manually annotate timestamps for 49 matches, establishing a more robust benchmark for soccer game commentary generation, termed as SN-Caption-test-align; Second, we propose a multi-modal temporal alignment pipeline to automatically correct and filter the existing dataset at scale, creating a higher-quality soccer game commentary dataset for training, denoted as MatchTime; Third, based on our curated dataset, we train an automatic commentary generation model, named MatchVoice. Extensive experiments and ablation studies have demonstrated the effectiveness of our alignment pipeline, and training model on the curated dataset achieves state-of-the-art performance for commentary generation, showcasing that better alignment can lead to significant performance improvements in downstream tasks." }, "2406.19852.pdf": { "id": "2406.19852v1", "title": "FootBots: A Transformer-based Architecture for Motion Prediction in Soccer", "year": "2024", "authors": [ "Guillem Capellera", "Luis Ferraz", "Antonio Rubio", "Antonio Agudo", "Francesc Moreno-Noguer" ], "categories": "cs.CV, cs.MA", "pdf_url": "https://arxiv.org/pdf/2406.19852v1", "abstract": "Motion prediction in soccer involves capturing complex dynamics from player and ball interactions. We present FootBots, an encoder-decoder transformer-based architecture addressing motion prediction and conditioned motion prediction through equivariance properties. FootBots captures temporal and social dynamics using set attention blocks and multi-attention block decoder. Our evaluation utilizes two datasets: a real soccer dataset and a tailored synthetic one. Insights from the synthetic dataset highlight the effectiveness of FootBots' social attention mechanism and the significance of conditioned motion prediction. Empirical results on real soccer data demonstrate that FootBots outperforms baselines in motion prediction and excels in conditioned tasks, such as predicting the players based on the ball position, predicting the offensive (defensive) team based on the ball and the defensive (offensive) team, and predicting the ball position based on all players. Our evaluation connects quantitative and qualitative findings. https://youtu.be/9kaEkfzG3L8" }, "2407.03041.pdf": { "id": "2407.03041v1", "title": "Position and Altitude of the Nao Camera Head from Two Points on the Soccer Field plus the Gravitational Direction", "year": "2024", "authors": [ "Stijn Oomes", "Arnoud Visser" ], "categories": "cs.RO, cs.CV", "pdf_url": "https://arxiv.org/pdf/2407.03041v1", "abstract": "To be able to play soccer, a robot needs a good estimate of its current position on the field. Ideally, multiple features are visible that have known locations. By applying trigonometry we can estimate the viewpoint from where this observation was actually made. Given that the Nao robots of the Standard Platform League have quite a limited field of view, a given camera frame typically only allows for one or two points to be recognized. In this paper we propose a method for determining the (x, y) coordinates on the field and the height h of the camera from the geometry of a simplified tetrahedron. This configuration is formed by two observed points on the ground plane plus the gravitational direction. When the distance between the two points is known, and the directions to the points plus the gravitational direction are measured, all dimensions of the tetrahedron can be determined. By performing these calculations with rational trigonometry instead of classical trigonometry, the computations turn out to be 28.7% faster, with equal numerical accuracy. The position of the head of the Nao can also be externally measured with the OptiTrack system. The difference between externally measured and internally predicted position from sensor data gives us mean absolute errors in the 3-6 centimeters range, when we estimated the gravitational direction from the vanishing point of the outer edges of the goal posts." }, "2407.03672.pdf": { "id": "2407.03672v1", "title": "A Survey of Data Synthesis Approaches", "year": "2024", "authors": [ "Hsin-Yu Chang", "Pei-Yu Chen", "Tun-Hsiang Chou", "Chang-Sheng Kao", "Hsuan-Yun Yu", "Yen-Ting Lin" ], "categories": "cs.LG, cs.AI", "pdf_url": "https://arxiv.org/pdf/2407.03672v1", "abstract": "This paper provides a detailed survey of synthetic data techniques. We first discuss the expected goals of using synthetic data in data augmentation, which can be divided into four parts: 1) Improving Diversity, 2) Data Balancing, 3) Addressing Domain Shift, and 4) Resolving Edge Cases. Synthesizing data are closely related to the prevailing machine learning techniques at the time, therefore, we summarize the domain of synthetic data techniques into four categories: 1) Expert-knowledge, 2) Direct Training, 3) Pre-train then Fine-tune, and 4) Foundation Models without Fine-tuning. Next, we categorize the goals of synthetic data filtering into four types for discussion: 1) Basic Quality, 2) Label Consistency, and 3) Data Distribution. In section 5 of this paper, we also discuss the future directions of synthetic data and state three direction that we believe is important: 1) focus more on quality, 2) the evaluation of synthetic data, and 3) multi-model data augmentation." }, "2407.08200.pdf": { "id": "2407.08200v1", "title": "Deep Understanding of Soccer Match Videos", "year": "2024", "authors": [ "Shikun Xu", "Yandong Zhu", "Gen Li", "Changhu Wang" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2407.08200v1", "abstract": "Soccer is one of the most popular sport worldwide, with live broadcasts frequently available for major matches. However, extracting detailed, frame-by-frame information on player actions from these videos remains a challenge. Utilizing state-of-the-art computer vision technologies, our system can detect key objects such as soccer balls, players and referees. It also tracks the movements of players and the ball, recognizes player numbers, classifies scenes, and identifies highlights such as goal kicks. By analyzing live TV streams of soccer matches, our system can generate highlight GIFs, tactical illustrations, and diverse summary graphs of ongoing games. Through these visual recognition techniques, we deliver a comprehensive understanding of soccer game videos, enriching the viewer's experience with detailed and insightful analysis." }, "2407.12483.pdf": { "id": "2407.12483v2", "title": "Towards AI-Powered Video Assistant Referee System (VARS) for Association Football", "year": "2024", "authors": [ "Jan Held", "Anthony Cioppa", "Silvio Giancola", "Abdullah Hamdi", "Christel Devue", "Bernard Ghanem" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2407.12483v2", "abstract": "Over the past decade, the technology used by referees in football has improved substantially, enhancing the fairness and accuracy of decisions. This progress has culminated in the implementation of the Video Assistant Referee (VAR), an innovation that enables backstage referees to review incidents on the pitch from multiple points of view. However, the VAR is currently limited to professional leagues due to its expensive infrastructure and the lack of referees worldwide. In this paper, we present the semi-automated Video Assistant Referee System (VARS) that leverages the latest findings in multi-view video analysis. VARS sets a new state-of-the-art on the SoccerNet-MVFoul dataset, a multi-view video dataset of football fouls. Our VARS achieves a new state-of-the-art on the SoccerNet-MVFoul dataset by recognizing the type of foul in 50% of instances and the appropriate sanction in 46% of cases. Finally, we conducted a comparative study to investigate human performance in classifying fouls and their corresponding severity and compared these findings to our VARS. The results of our study highlight the potential of our VARS to reach human performance and support football refereeing across all levels of professional and amateur federations." }, "2407.13109.pdf": { "id": "2407.13109v1", "title": "A Framework for Spatio-Temporal Graph Analytics In Field Sports", "year": "2024", "authors": [ "Valerio Antonini", "Michael Scriney", "Alessandra Mileo", "Mark Roantree" ], "categories": "cs.CY, cs.LG", "pdf_url": "https://arxiv.org/pdf/2407.13109v1", "abstract": "The global sports analytics industry has a market value of USD 3.78 billion in 2023. The increase of wearables such as GPS sensors has provided analysts with large fine-grained datasets detailing player performance. Traditional analysis of this data focuses on individual athletes with measures of internal and external loading such as distance covered in speed zones or rate of perceived exertion. However these metrics do not provide enough information to understand team dynamics within field sports. The spatio-temporal nature of match play necessitates an investment in date-engineering to adequately transform the data into a suitable format to extract features such as areas of activity. In this paper we present an approach to construct Time-Window Spatial Activity Graphs (TWGs) for field sports. Using GPS data obtained from Gaelic Football matches we demonstrate how our approach can be utilised to extract spatio-temporal features from GPS sensor data" }, "2407.14120.pdf": { "id": "2407.14120v1", "title": "The Cardinality of Identifying Code Sets for Soccer Ball Graph with Application to Remote Sensing", "year": "2024", "authors": [ "Anna L. D. Latour", "Arunabha Sen", "Kaustav Basu", "Chenyang Zhou", "Kuldeep S. Meel" ], "categories": "cs.AI", "pdf_url": "https://arxiv.org/pdf/2407.14120v1", "abstract": "In the context of satellite monitoring of the earth, we can assume that the surface of the earth is divided into a set of regions. We assume that the impact of a big social/environmental event spills into neighboring regions. Using Identifying Code Sets (ICSes), we can deploy sensors in such a way that the region in which an event takes place can be uniquely identified, even with fewer sensors than regions. As Earth is almost a sphere, we use a soccer ball as a model. We construct a Soccer Ball Graph (SBG), and provide human-oriented, analytical proofs that 1) the SBG has at least 26 ICSes of cardinality ten, implying that there are at least 26 different ways to deploy ten satellites to monitor the Earth and 2) that the cardinality of the minimum Identifying Code Set (MICS) for the SBG is at least nine. We then provide a machine-oriented formal proof that the cardinality of the MICS for the SBG is in fact ten, meaning that one must deploy at least ten satellites to monitor the Earth in the SBG model. We also provide machine-oriented proof that there are exactly 26 ICSes of cardinality ten for the SBG." }, "2407.14516.pdf": { "id": "2407.14516v1", "title": "RobocupGym: A challenging continuous control benchmark in Robocup", "year": "2024", "authors": [ "Michael Beukman", "Branden Ingram", "Geraud Nangue Tasse", "Benjamin Rosman", "Pravesh Ranchod" ], "categories": "cs.RO, cs.LG", "pdf_url": "https://arxiv.org/pdf/2407.14516v1", "abstract": "Reinforcement learning (RL) has progressed substantially over the past decade, with much of this progress being driven by benchmarks. Many benchmarks are focused on video or board games, and a large number of robotics benchmarks lack diversity and real-world applicability. In this paper, we aim to simplify the process of applying reinforcement learning in the 3D simulation league of Robocup, a robotic football competition. To this end, we introduce a Robocup-based RL environment based on the open source rcssserver3d soccer server, simple pre-defined tasks, and integration with a popular RL library, Stable Baselines 3. Our environment enables the creation of high-dimensional continuous control tasks within a robotics football simulation. In each task, an RL agent controls a simulated Nao robot, and can interact with the ball or other agents. We open-source our environment and training code at https://github.com/Michael-Beukman/RobocupGym." }, "2407.14558.pdf": { "id": "2407.14558v1", "title": "A Foundation Model for Soccer", "year": "2024", "authors": [ "Ethan Baron", "Daniel Hocevar", "Zach Salehe" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2407.14558v1", "abstract": "We propose a foundation model for soccer, which is able to predict subsequent actions in a soccer match from a given input sequence of actions. As a proof of concept, we train a transformer architecture on three seasons of data from a professional soccer league. We quantitatively and qualitatively compare the performance of this transformer architecture to two baseline models: a Markov model and a multi-layer perceptron. Additionally, we discuss potential applications of our model. We provide an open-source implementation of our methods at https://github.com/danielhocevar/Foundation-Model-for-Soccer." }, "2407.14612.pdf": { "id": "2407.14612v1", "title": "A Biomechanics-Inspired Approach to Soccer Kicking for Humanoid Robots", "year": "2024", "authors": [ "Daniel Marew", "Nisal Perera", "Shangqun Yu", "Sarah Roelker", "Donghyun Kim" ], "categories": "cs.RO", "pdf_url": "https://arxiv.org/pdf/2407.14612v1", "abstract": "Soccer kicking is a complex whole-body motion that requires intricate coordination of various motor actions. To accomplish such dynamic motion in a humanoid robot, the robot needs to simultaneously: 1) transfer high kinetic energy to the kicking leg, 2) maintain balance and stability of the entire body, and 3) manage the impact disturbance from the ball during the kicking moment. Prior studies on robotic soccer kicking often prioritized stability, leading to overly conservative quasi-static motions. In this work, we present a biomechanics-inspired control framework that leverages trajectory optimization and imitation learning to facilitate highly dynamic soccer kicks in humanoid robots. We conducted an in-depth analysis of human soccer kick biomechanics to identify key motion constraints. Based on this understanding, we designed kinodynamically feasible trajectories that are then used as a reference in imitation learning to develop a robust feedback control policy. We demonstrate the effectiveness of our approach through a simulation of an anthropomorphic 25 DoF bipedal humanoid robot, named PresToe, which is equipped with 7 DoF legs, including a unique actuated toe. Using our framework, PresToe can execute dynamic instep kicks, propelling the ball at speeds exceeding 11m/s in full dynamics simulation." }, "2407.17832.pdf": { "id": "2407.17832v2", "title": "Regularized Adjusted Plus-Minus Models for Evaluating and Scouting Football (Soccer) Players using Possession Sequences", "year": "2024", "authors": [ "Robert Bajons", "Kurt Hornik" ], "categories": "stat.AP", "pdf_url": "https://arxiv.org/pdf/2407.17832v2", "abstract": "This paper presents a novel framework for evaluating players in association football (soccer). Our method uses possession sequences, i.e. sequences of consecutive on-ball actions, for deriving estimates for player strengths. On the surface, the methodology is similar to classical adjusted plus-minus rating models using mainly regularized regression techniques. However, by analyzing possessions, our framework is able to distinguish on-ball and off-ball contributions of players to the game. From a methodological viewpoint, the framework explores four different penalization schemes, which exploit football-specific structures such as the grouping of players into position groups as well as into common strength groups. These four models lead to four ways to rate players by considering the respective estimate of each model corresponding to the player. The ratings are used to analyze the 2017/18 season of the Spanish La Liga. We compare similarities as well as particular use cases of each of the penalized models and provide guidance for practitioners when using the individual model specifications. Finally, we conclude our analysis by providing a domain-specific statistical evaluation framework, which highlights the potential of the penalized regression approaches for evaluating players." }, "2407.19285.pdf": { "id": "2407.19285v1", "title": "The Impact of Foreign Players in the English Premier League: A Mathematical Analys", "year": "2024", "authors": [ "Amit K Chattopadhyay", "A. Abdul", "Sudhir Jain" ], "categories": "stat.OT, cond-mat.stat-mech", "pdf_url": "https://arxiv.org/pdf/2407.19285v1", "abstract": "We undertake extensive analysis of English Premier League data over the period 2009/10 to 2017/18 to identify and rank key factors affecting the economic and footballing performances of the teams. Alternative end-of-season league tables are generated by re-ranking the teams based on five different descriptors - total expenditure, total funds spent on players, total funds spent on foreign players, the ratio of foreign to British players and the overall profit. The unequal distribution of resources and expenditure between the clubs is analyzed through Lorenz curves. A comparative analysis of the differences between the alternative tables and the conventional end-of-season league table establishes the most likely factors to influence the performances of the teams that we also rank using Principal Component Analysis. We find that the top teams in the league are also those that tend to have the highest expenditure overall, for all players, including foreign players; they also have the highest ratios of foreign to British players. Our statistical and machine learning study also indicates that successful performance on the field may not guarantee healthy profits at the end of the season." }, "2407.19686.pdf": { "id": "2407.19686v1", "title": "Billiards Sports Analytics: Datasets and Tasks", "year": "2024", "authors": [ "Qianru Zhang", "Zheng Wang", "Cheng Long", "Siu-Ming Yiu" ], "categories": "cs.CE", "pdf_url": "https://arxiv.org/pdf/2407.19686v1", "abstract": "Nowadays, it becomes a common practice to capture some data of sports games with devices such as GPS sensors and cameras and then use the data to perform various analyses on sports games, including tactics discovery, similar game retrieval, performance study, etc. While this practice has been conducted to many sports such as basketball and soccer, it remains largely unexplored on the billiards sports, which is mainly due to the lack of publicly available datasets. Motivated by this, we collect a dataset of billiards sports, which includes the layouts (i.e., locations) of billiards balls after performing break shots, called break shot layouts, the traces of the balls as a result of strikes (in the form of trajectories), and detailed statistics and performance indicators. We then study and develop techniques for three tasks on the collected dataset, including (1) prediction and (2) generation on the layouts data, and (3) similar billiards layout retrieval on the layouts data, which can serve different users such as coaches, players and fans. We conduct extensive experiments on the collected dataset and the results show that our methods perform effectively and efficiently." }, "2408.02146.pdf": { "id": "2408.02146v1", "title": "Video-based Pedestrian and Vehicle Traffic Analysis During Football Games", "year": "2024", "authors": [ "Jacques P. Fleischer", "Ryan Pallack", "Ahan Mishra", "Gustavo Riente de Andrade", "Subhadipto Poddar", "Emmanuel Posadas" ], "categories": "cs.CV, cs.CY", "pdf_url": "https://arxiv.org/pdf/2408.02146v1", "abstract": "This paper utilizes video analytics to study pedestrian and vehicle traffic behavior, focusing on analyzing traffic patterns during football gamedays. The University of Florida (UF) hosts six to seven home football games on Saturdays during the college football season, attracting significant pedestrian activity. Through video analytics, this study provides valuable insights into the impact of these events on traffic volumes and safety at intersections. Comparing pedestrian and vehicle activities on gamedays versus non-gamedays reveals differing patterns. For example, pedestrian volume substantially increases during gamedays, which is positively correlated with the probability of the away team winning. This correlation is likely because fans of the home team enjoy watching difficult games. Win probabilities as an early predictor of pedestrian volumes at intersections can be a tool to help traffic professionals anticipate traffic management needs. Pedestrian-to-vehicle (P2V) conflicts notably increase on gamedays, particularly a few hours before games start. Addressing this, a \"Barnes Dance\" movement phase within the intersection is recommended. Law enforcement presence during high-activity gamedays can help ensure pedestrian compliance and enhance safety. In contrast, we identified that vehicle-to-vehicle (V2V) conflicts generally do not increase on gamedays and may even decrease due to heightened driver caution." }, "2408.07927.pdf": { "id": "2408.07927v1", "title": "Motif analysis and passing behavior in football passing networks", "year": "2024", "authors": [ "Ming-Xia Li", "Li-Gong Xu", "Wei-Xing Zhou" ], "categories": "physics.soc-ph", "pdf_url": "https://arxiv.org/pdf/2408.07927v1", "abstract": "The strategic orchestration of football matchplays profoundly influences game outcomes, motivating a surge in research aimed at uncovering tactical nuances through social network analysis. In this paper, we delve into the microscopic intricacies of cooperative player interactions by focusing on triadic motifs within passing networks. Employing a dataset compiled from 3,199 matches across 18 premier football competitions, we identify successful passing activities and construct passing networks for both home and away teams. Our findings highlight a pronounced disparity in passing efficiency, with home teams demonstrating superior performance relative to away teams. Through the identification and analysis of 3-motifs, we find that the motifs with more bidirectional links are more significant. It reveals that footballers exhibit a strong tendency towards backward passes rather than direct forward attacks. Comparing the results of games, we find that some motifs are related to the goal difference. It indicates that direct and effective forward passing significantly amplifies a team's offensive capabilities, whereas an abundance of passbacks portends an elevated risk of offensive futility. These revelations affirm the efficacy of network motif analysis as a potent analytical tool for unveiling the foundational components of passing dynamics among footballers and for decoding the complex tactical behaviors and interaction modalities that underpin team performance." }, "2408.08331.pdf": { "id": "2408.08331v1", "title": "Match predictions in soccer: Machine learning vs. Poisson approaches", "year": "2024", "authors": [ "Mirko Fischer", "Andreas Heuer" ], "categories": "stat.AP", "pdf_url": "https://arxiv.org/pdf/2408.08331v1", "abstract": "Predicting the results of soccer matches is of great interest. This is not only due to the popularity of the sport and the joy of private \"betting rounds\", but also due to the large sports betting market. Where previously expert knowledge and intuition were used, today there are models that analyze large amounts of data and make predictions based on them. In addition to Poisson models, approaches that belong to the machine learning (ML) category are increasingly being used. These include, for example, neural network or random forest models, which are compared in this article with each other as well as with Poisson models with regard to single-match prediction. In each case, the match results of a season are used as the data basis. The analysis is carried out for 5 European top leagues. A statistical analysis shows that the performance levels of the teams do not change systematically during a season. In order to characterize performance levels as accurately as possible, all match results, except from the match to be predicted, can be used as features with equal weighting. It can be seen that both, the exact choice of features and the choice of model, have only a minor influence on the prediction quality. Possible improvements in match prediction are discussed." }, "2409.13098.pdf": { "id": "2409.13098v1", "title": "Predicting soccer matches with complex networks and machine learning", "year": "2024", "authors": [ "Eduardo Alves Baratela", "Felipe Jord\u00e3o Xavier", "Thomas Peron", "Paulino Ribeiro Villas-Boas", "Francisco Aparecido Rodrigues" ], "categories": "cs.SI, cs.LG, stat.AP, stat.ML", "pdf_url": "https://arxiv.org/pdf/2409.13098v1", "abstract": "Soccer attracts the attention of many researchers and professionals in the sports industry. Therefore, the incorporation of science into the sport is constantly growing, with increasing investments in performance analysis and sports prediction industries. This study aims to (i) highlight the use of complex networks as an alternative tool for predicting soccer match outcomes, and (ii) show how the combination of structural analysis of passing networks with match statistical data can provide deeper insights into the game patterns and strategies used by teams. In order to do so, complex network metrics and match statistics were used to build machine learning models that predict the wins and losses of soccer teams in different leagues. The results showed that models based on passing networks were as effective as ``traditional'' models, which use general match statistics. Another finding was that by combining both approaches, more accurate models were obtained than when they were used separately, demonstrating that the fusion of such approaches can offer a deeper understanding of game patterns, allowing the comprehension of tactics employed by teams relationships between players, their positions, and interactions during matches. It is worth mentioning that both network metrics and match statistics were important and impactful for the mixed model. Furthermore, the use of networks with a lower granularity of temporal evolution (such as creating a network for each half of the match) performed better than a single network for the entire game." }, "2409.13859.pdf": { "id": "2409.13859v1", "title": "PanoCoach: Enhancing Tactical Coaching and Communication in Soccer with Mixed-Reality Telepresence", "year": "2024", "authors": [ "Andrew Kang", "Hanspeter Pfister", "Tica Lin" ], "categories": "cs.HC, cs.GR", "pdf_url": "https://arxiv.org/pdf/2409.13859v1", "abstract": "Soccer, as a dynamic team sport, requires seamless coordination and integration of tactical strategies across all players. Adapting to new tactical systems is a critical but often challenging aspect of soccer at all professional levels. Even the best players can struggle with this process, primarily due to the complexities of conveying and internalizing intricate tactical patterns. Traditional communication methods like whiteboards, on-field instructions, and video analysis often present significant difficulties in perceiving spatial relationships, anticipating team movements, and facilitating live conversation during training sessions. These challenges can lead to inconsistent interpretations of the coach's tactics among players, regardless of their skill level. To bridge the gap between tactical communication and physical execution, we propose a mixed-reality telepresence solution designed to support multi-view tactical explanations during practice. Our concept involves a multi-screen setup combining a tablet for coaches to annotate and demonstrate concepts in both 2D and 3D views, alongside VR to immerse athletes in a first-person perspective, allowing them to experience a sense of presence during coaching. Demo video uploaded at https://youtu.be/O7o4Wzd-7rw" }, "2409.14550.pdf": { "id": "2409.14550v1", "title": "Interpretable Nonroutine Network Traffic Prediction with a Case Study", "year": "2024", "authors": [ "Liangzhi Wang", "Haoyuan Zhu", "Jiliang Zhang", "Zitian Zhang", "Jie Zhang" ], "categories": "cs.NI", "pdf_url": "https://arxiv.org/pdf/2409.14550v1", "abstract": "This paper pioneers a nonroutine network traffic prediction (NNTP) method to prospectively provide a theoretical basis for avoiding large-scale network disruption by accurately predicting bursty traffic. Certain events that impact user behavior subsequently trigger nonroutine traffic, which significantly constrains the performance of network traffic prediction (NTP) models. By analyzing nonroutine traffic and the corresponding events, the NNTP method is pioneered to construct interpretable NTP model. Based on the real-world traffic data, the network traffic generated during soccer games serves as a case study to validate the performance of the NNTP method. The numerical results indicate that our prediction closely fits the traffic pattern. In comparison to existing researches, the NNTP method is at the forefront of finding a balance among interpretability, accuracy, and computational complexity." }, "2410.00943.pdf": { "id": "2410.00943v1", "title": "RisingBALLER: A player is a token, a match is a sentence, A path towards a foundational model for football players data analytics", "year": "2024", "authors": [ "Akedjou Achraff Adjileye" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2410.00943v1", "abstract": "In this paper, I introduce RisingBALLER, the first publicly available approach that leverages a transformer model trained on football match data to learn match-specific player representations. Drawing inspiration from advances in language modeling, RisingBALLER treats each football match as a unique sequence in which players serve as tokens, with their embeddings shaped by the specific context of the match. Through the use of masked player prediction (MPP) as a pre-training task, RisingBALLER learns foundational features for football player representations, similar to how language models learn semantic features for text representations. As a downstream task, I introduce next match statistics prediction (NMSP) to showcase the effectiveness of the learned player embeddings. The NMSP model surpasses a strong baseline commonly used for performance forecasting within the community. Furthermore, I conduct an in-depth analysis to demonstrate how the learned embeddings by RisingBALLER can be used in various football analytics tasks, such as producing meaningful positional features that capture the essence and variety of player roles beyond rigid x,y coordinates, team cohesion estimation, and similar player retrieval for more effective data-driven scouting. More than a simple machine learning model, RisingBALLER is a comprehensive framework designed to transform football data analytics by learning high-level foundational features for players, taking into account the context of each match. It offers a deeper understanding of football players beyond individual statistics." }, "2410.01304.pdf": { "id": "2410.01304v1", "title": "Deep learning for action spotting in association football videos", "year": "2024", "authors": [ "Silvio Giancola", "Anthony Cioppa", "Bernard Ghanem", "Marc Van Droogenbroeck" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2410.01304v1", "abstract": "The task of action spotting consists in both identifying actions and precisely localizing them in time with a single timestamp in long, untrimmed video streams. Automatically extracting those actions is crucial for many sports applications, including sports analytics to produce extended statistics on game actions, coaching to provide support to video analysts, or fan engagement to automatically overlay content in the broadcast when specific actions occur. However, before 2018, no large-scale datasets for action spotting in sports were publicly available, which impeded benchmarking action spotting methods. In response, our team built the largest dataset and the most comprehensive benchmarks for sports video understanding, under the umbrella of SoccerNet. Particularly, our dataset contains a subset specifically dedicated to action spotting, called SoccerNet Action Spotting, containing more than 550 complete broadcast games annotated with almost all types of actions that can occur in a football game. This dataset is tailored to develop methods for automatic spotting of actions of interest, including deep learning approaches, by providing a large amount of manually annotated actions. To engage with the scientific community, the SoccerNet initiative organizes yearly challenges, during which participants from all around the world compete to achieve state-of-the-art performances. Thanks to our dataset and challenges, more than 60 methods were developed or published over the past five years, improving on the first baselines and making action spotting a viable option for the sports industry. This paper traces the history of action spotting in sports, from the creation of the task back in 2018, to the role it plays today in research and the sports industry." }, "2410.07401.pdf": { "id": "2410.07401v1", "title": "Enhancing Soccer Camera Calibration Through Keypoint Exploitation", "year": "2024", "authors": [ "Nikolay S. Falaleev", "Ruilong Chen" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2410.07401v1", "abstract": "Accurate camera calibration is essential for transforming 2D images from camera sensors into 3D world coordinates, enabling precise scene geometry interpretation and supporting sports analytics tasks such as player tracking, offside detection, and performance analysis. However, obtaining a sufficient number of high-quality point pairs remains a significant challenge for both traditional and deep learning-based calibration methods. This paper introduces a multi-stage pipeline that addresses this challenge by leveraging the structural features of the football pitch. Our approach significantly increases the number of usable points for calibration by exploiting line-line and line-conic intersections, points on the conics, and other geometric features. To mitigate the impact of imperfect annotations, we employ data fitting techniques. Our pipeline utilizes deep learning for keypoint and line detection and incorporates geometric constraints based on real-world pitch dimensions. A voter algorithm iteratively selects the most reliable keypoints, further enhancing calibration accuracy. We evaluated our approach on the largest football broadcast camera calibration dataset available, and secured the top position in the SoccerNet Camera Calibration Challenge 2023 [arXiv:2309.06006], which demonstrates the effectiveness of our method in real-world scenarios. The project code is available at https://github.com/NikolasEnt/soccernet-calibration-sportlight ." }, "2410.08931.pdf": { "id": "2410.08931v1", "title": "Enhancing Motion Variation in Text-to-Motion Models via Pose and Video Conditioned Editing", "year": "2024", "authors": [ "Clayton Leite", "Yu Xiao" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2410.08931v1", "abstract": "Text-to-motion models that generate sequences of human poses from textual descriptions are garnering significant attention. However, due to data scarcity, the range of motions these models can produce is still limited. For instance, current text-to-motion models cannot generate a motion of kicking a football with the instep of the foot, since the training data only includes martial arts kicks. We propose a novel method that uses short video clips or images as conditions to modify existing basic motions. In this approach, the model's understanding of a kick serves as the prior, while the video or image of a football kick acts as the posterior, enabling the generation of the desired motion. By incorporating these additional modalities as conditions, our method can create motions not present in the training set, overcoming the limitations of text-motion datasets. A user study with 26 participants demonstrated that our approach produces unseen motions with realism comparable to commonly represented motions in text-motion datasets (e.g., HumanML3D), such as walking, running, squatting, and kicking." }, "2410.09068.pdf": { "id": "2410.09068v1", "title": "Modeling and Prediction of the UEFA EURO 2024 via Combined Statistical Learning Approaches", "year": "2024", "authors": [ "Andreas Groll", "Lars M. Hvattum", "Christophe Ley", "Jonas Sternemann", "Gunther Schauberger", "Achim Zeileis" ], "categories": "cs.LG, stat.AP", "pdf_url": "https://arxiv.org/pdf/2410.09068v1", "abstract": "In this work, three fundamentally different machine learning models are combined to create a new, joint model for forecasting the UEFA EURO 2024. Therefore, a generalized linear model, a random forest model, and a extreme gradient boosting model are used to predict the number of goals a team scores in a match. The three models are trained on the match results of the UEFA EUROs 2004-2020, with additional covariates characterizing the teams for each tournament as well as three enhanced variables derived from different ranking methods for football teams. The first enhanced variable is based on historic match data from national teams, the second is based on the bookmakers' tournament winning odds of all participating teams, and the third is based on historic match data of individual players both for club and international matches, resulting in player ratings. Then, based on current covariate information of the participating teams, the final trained model is used to predict the UEFA EURO 2024. For this purpose, the tournament is simulated 100.000 times, based on the estimated expected number of goals for all possible matches, from which probabilities across the different tournament stages are derived. Our combined model identifies France as the clear favourite with a winning probability of 19.2%, followed by England (16.7%) and host Germany (13.7%)." }, "2410.17785.pdf": { "id": "2410.17785v2", "title": "TranSPORTmer: A Holistic Approach to Trajectory Understanding in Multi-Agent Sports", "year": "2024", "authors": [ "Guillem Capellera", "Luis Ferraz", "Antonio Rubio", "Antonio Agudo", "Francesc Moreno-Noguer" ], "categories": "cs.CV, cs.MA", "pdf_url": "https://arxiv.org/pdf/2410.17785v2", "abstract": "Understanding trajectories in multi-agent scenarios requires addressing various tasks, including predicting future movements, imputing missing observations, inferring the status of unseen agents, and classifying different global states. Traditional data-driven approaches often handle these tasks separately with specialized models. We introduce TranSPORTmer, a unified transformer-based framework capable of addressing all these tasks, showcasing its application to the intricate dynamics of multi-agent sports scenarios like soccer and basketball. Using Set Attention Blocks, TranSPORTmer effectively captures temporal dynamics and social interactions in an equivariant manner. The model's tasks are guided by an input mask that conceals missing or yet-to-be-predicted observations. Additionally, we introduce a CLS extra agent to classify states along soccer trajectories, including passes, possessions, uncontrolled states, and out-of-play intervals, contributing to an enhancement in modeling trajectories. Evaluations on soccer and basketball datasets show that TranSPORTmer outperforms state-of-the-art task-specific models in player forecasting, player forecasting-imputation, ball inference, and ball imputation. https://youtu.be/8VtSRm8oGoE" }, "2410.20038.pdf": { "id": "2410.20038v1", "title": "Revisiting PlayeRank", "year": "2024", "authors": [ "Louise Schmidt", "Cristian Lillo", "Javier Bustos" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2410.20038v1", "abstract": "In this article we revise the football's performance score called PlayeRank, designed and evaluated by Pappalardo et al.\\ in 2019. First, we analyze the weights extracted from the Linear Support Vector Machine (SVM) that solves the classification problem of \"which set of events has a higher impact on the chances of winning a match\". Here, we notice that the previously published results include the Goal-Scored event during the training phase, which produces inconsistencies. We fix these inconsistencies, and show new weights capable of solving the same problem. Following the intuition that the best team should always win a match, we define the team's quality as the average number of players involved in the game. We show that, using the original PlayeRank, in 94.13\\% of the matches either the superior team beats the inferior team or the teams end tied if the scores are similar. Finally, we present a way to use PlayeRank in an online fashion using modified free analysis tools. Calculating this modified version of PlayeRank, we performed an online analysis of a real football match every five minutes of game. Here, we evaluate the usefulness of that information with experts and managers, and conclude that the obtained data indeed provides useful information that was not previously available to the manager during the match." }, "2410.21484.pdf": { "id": "2410.21484v1", "title": "A Systematic Review of Machine Learning in Sports Betting: Techniques, Challenges, and Future Directions", "year": "2024", "authors": [ "Ren\u00e9 Manass\u00e9 Galekwa", "Jean Marie Tshimula", "Etienne Gael Tajeuna", "Kyamakya Kyandoghere" ], "categories": "cs.LG, cs.CE, cs.ET, cs.IR, cs.SI", "pdf_url": "https://arxiv.org/pdf/2410.21484v1", "abstract": "The sports betting industry has experienced rapid growth, driven largely by technological advancements and the proliferation of online platforms. Machine learning (ML) has played a pivotal role in the transformation of this sector by enabling more accurate predictions, dynamic odds-setting, and enhanced risk management for both bookmakers and bettors. This systematic review explores various ML techniques, including support vector machines, random forests, and neural networks, as applied in different sports such as soccer, basketball, tennis, and cricket. These models utilize historical data, in-game statistics, and real-time information to optimize betting strategies and identify value bets, ultimately improving profitability. For bookmakers, ML facilitates dynamic odds adjustment and effective risk management, while bettors leverage data-driven insights to exploit market inefficiencies. This review also underscores the role of ML in fraud detection, where anomaly detection models are used to identify suspicious betting patterns. Despite these advancements, challenges such as data quality, real-time decision-making, and the inherent unpredictability of sports outcomes remain. Ethical concerns related to transparency and fairness are also of significant importance. Future research should focus on developing adaptive models that integrate multimodal data and manage risk in a manner akin to financial portfolios. This review provides a comprehensive examination of the current applications of ML in sports betting, and highlights both the potential and the limitations of these technologies." }, "2411.08901.pdf": { "id": "2411.08901v1", "title": "SoccerGuard: Investigating Injury Risk Factors for Professional Soccer Players with Machine Learning", "year": "2024", "authors": [ "Finn Bartels", "Lu Xing", "Cise Midoglu", "Matthias Boeker", "Toralf Kirsten", "P\u00e5l Halvorsen" ], "categories": "cs.HC, cs.LG", "pdf_url": "https://arxiv.org/pdf/2411.08901v1", "abstract": "We present SoccerGuard, a novel framework for predicting injuries in women's soccer using Machine Learning (ML). This framework can ingest data from multiple sources, including subjective wellness and training load reports from players, objective GPS sensor measurements, third-party player statistics, and injury reports verified by medical personnel. We experiment with a number of different settings related to synthetic data generation, input and output window sizes, and ML models for prediction. Our results show that, given the right configurations and feature combinations, injury event prediction can be undertaken with considerable accuracy. The optimal results are achieved when input windows are reduced and larger combined output windows are defined, in combination with an ideally balanced data set. The framework also includes a dashboard with a user-friendly Graphical User Interface (GUI) to support interactive analysis and visualization." }, "2411.09085.pdf": { "id": "2411.09085v1", "title": "Predictive Modeling of Lower-Level English Club Soccer Using Crowd-Sourced Player Valuations", "year": "2024", "authors": [ "Josh Brown", "Yutong Bu", "Zachary Cheesman", "Benjamin Orman", "Iris Horng", "Samuel Thomas" ], "categories": "stat.AP", "pdf_url": "https://arxiv.org/pdf/2411.09085v1", "abstract": "In this research, we examine the capabilities of different mathematical models to accurately predict various levels of the English football pyramid. Existing work has largely focused on top-level play in European leagues; however, our work analyzes teams throughout the entire English Football League system. We modeled team performance using weighted Colley and Massey ranking methods which incorporate player valuations from the widely-used website Transfermarkt to predict game outcomes. Our initial analysis found that lower leagues are more difficult to forecast in general. Yet, after removing dominant outlier teams from the analysis, we found that top leagues were just as difficult to predict as lower leagues. We also extended our findings using data from multiple German and Scottish leagues. Finally, we discuss reasons to doubt attributing Transfermarkt's predictive value to wisdom of the crowd." }, "2411.16216.pdf": { "id": "2411.16216v1", "title": "SMGDiff: Soccer Motion Generation using diffusion probabilistic models", "year": "2024", "authors": [ "Hongdi Yang", "Chengyang Li", "Zhenxuan Wu", "Gaozheng Li", "Jingya Wang", "Jingyi Yu" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2411.16216v1", "abstract": "Soccer is a globally renowned sport with significant applications in video games and VR/AR. However, generating realistic soccer motions remains challenging due to the intricate interactions between the human player and the ball. In this paper, we introduce SMGDiff, a novel two-stage framework for generating real-time and user-controllable soccer motions. Our key idea is to integrate real-time character control with a powerful diffusion-based generative model, ensuring high-quality and diverse output motion. In the first stage, we instantly transform coarse user controls into diverse global trajectories of the character. In the second stage, we employ a transformer-based autoregressive diffusion model to generate soccer motions based on trajectory conditioning. We further incorporate a contact guidance module during inference to optimize the contact details for realistic ball-foot interactions. Moreover, we contribute a large-scale soccer motion dataset consisting of over 1.08 million frames of diverse soccer motions. Extensive experiments demonstrate that our SMGDiff significantly outperforms existing methods in terms of motion quality and condition alignment." }, "2411.17450.pdf": { "id": "2411.17450v2", "title": "A Graph Neural Network deep-dive into successful counterattacks", "year": "2024", "authors": [ "Joris Bekkers", "Amod Sahasrabudhe" ], "categories": "cs.LG, cs.SI", "pdf_url": "https://arxiv.org/pdf/2411.17450v2", "abstract": "A counterattack in soccer is a high speed, high intensity direct attack that can occur when a team transitions from a defensive state to an attacking state after regaining possession of the ball. The aim is to create a goal-scoring opportunity by convering a lot of ground with minimal passes before the opposing team can recover their defensive shape. The purpose of this research is to build gender-specific Graph Neural Networks to model the likelihood of a counterattack being successful and uncover what factors make them successful in professional soccer. These models are trained on a total of 20863 frames of synchronized on-ball event and spatiotemporal (broadcast) tracking data. This dataset is derived from 632 games of MLS (2022), NWSL (2022) and international soccer (2020-2022). With this data we demonstrate that gender-specific Graph Neural Networks outperform architecturally identical gender-ambiguous models in predicting the successful outcome of counterattacks. We show, using Permutation Feature Importance, that byline to byline speed, angle to the goal, angle to the ball and sideline to sideline speed are the node features with the highest impact on model performance. Additionally, we offer some illustrative examples on how to navigate the infinite solution search space to aid in identifying improvements for player decision making. This research is accompanied by an open-source repository containing all data and code, and it is also accompanied by an open-source Python package which simplifies converting spatiotemporal data into graphs. This package also facilitates testing, validation, training and prediction with this data. This should allow the reader to replicate and improve upon our research more easily." }, "2412.01721.pdf": { "id": "2412.01721v1", "title": "BroadTrack: Broadcast Camera Tracking for Soccer", "year": "2024", "authors": [ "Floriane Magera", "Thomas Hoyoux", "Olivier Barnich", "Marc Van Droogenbroeck" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2412.01721v1", "abstract": "Camera calibration and localization, sometimes simply named camera calibration, enables many applications in the context of soccer broadcasting, for instance regarding the interpretation and analysis of the game, or the insertion of augmented reality graphics for storytelling or refereeing purposes. To contribute to such applications, the research community has typically focused on single-view calibration methods, leveraging the near-omnipresence of soccer field markings in wide-angle broadcast views, but leaving all temporal aspects, if considered at all, to general-purpose tracking or filtering techniques. Only a few contributions have been made to leverage any domain-specific knowledge for this tracking task, and, as a result, there lacks a truly performant and off-the-shelf camera tracking system tailored for soccer broadcasting, specifically for elevated tripod-mounted cameras around the stadium. In this work, we present such a system capable of addressing the task of soccer broadcast camera tracking efficiently, robustly, and accurately, outperforming by far the most precise methods of the state-of-the-art. By combining the available open-source soccer field detectors with carefully designed camera and tripod models, our tracking system, BroadTrack, halves the mean reprojection error rate and gains more than 15% in terms of Jaccard index for camera calibration on the SoccerNet dataset. Furthermore, as the SoccerNet dataset videos are relatively short (30 seconds), we also present qualitative results on a 20-minute broadcast clip to showcase the robustness and the soundness of our system." }, "2412.01820.pdf": { "id": "2412.01820v3", "title": "Towards Universal Soccer Video Understanding", "year": "2024", "authors": [ "Jiayuan Rao", "Haoning Wu", "Hao Jiang", "Ya Zhang", "Yanfeng Wang", "Weidi Xie" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2412.01820v3", "abstract": "As a globally celebrated sport, soccer has attracted widespread interest from fans all over the world. This paper aims to develop a comprehensive multi-modal framework for soccer video understanding. Specifically, we make the following contributions in this paper: (i) we introduce SoccerReplay-1988, the largest multi-modal soccer dataset to date, featuring videos and detailed annotations from 1,988 complete matches, with an automated annotation pipeline; (ii) we present an advanced soccer-specific visual encoder, MatchVision, which leverages spatiotemporal information across soccer videos and excels in various downstream tasks; (iii) we conduct extensive experiments and ablation studies on event classification, commentary generation, and multi-view foul recognition. MatchVision demonstrates state-of-the-art performance on all of them, substantially outperforming existing models, which highlights the superiority of our proposed data and model. We believe that this work will offer a standard paradigm for sports understanding research." }, "2412.05911.pdf": { "id": "2412.05911v1", "title": "Unveiling True Talent: The Soccer Factor Model for Skill Evaluation", "year": "2024", "authors": [ "Alexandre Andorra", "Maximilian G\u00f6bel" ], "categories": "stat.AP, econ.GN", "pdf_url": "https://arxiv.org/pdf/2412.05911v1", "abstract": "Evaluating a soccer player's performance can be challenging due to the high costs and small margins involved in recruitment decisions. Raw observational statistics further complicate an accurate individual skill assessment as they do not abstract from the potentially confounding factor of team strength. We introduce the Soccer Factor Model (SFM), which corrects this bias by isolating a player's true skill from the team's influence. We compile a novel data set, web-scraped from publicly available data sources. Our empirical application draws on information of 144 players, playing a total of over 33,000 matches, in seasons 2000/01 through 2023/24. Not only does the SFM allow for a structural interpretation of a player's skill, but also stands out against more reduced-form benchmarks in terms of forecast accuracy. Moreover, we propose Skill- and Performance Above Replacement as metrics for fair cross-player comparisons. These, for example, allow us to settle the discussion about the GOAT of soccer in the first quarter of the twenty-first century." }, "2412.15517.pdf": { "id": "2412.15517v1", "title": "Novelty-Guided Data Reuse for Efficient and Diversified Multi-Agent Reinforcement Learning", "year": "2024", "authors": [ "Yangkun Chen", "Kai Yang", "Jian Tao", "Jiafei Lyu" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2412.15517v1", "abstract": "Recently, deep Multi-Agent Reinforcement Learning (MARL) has demonstrated its potential to tackle complex cooperative tasks, pushing the boundaries of AI in collaborative environments. However, the efficiency of these systems is often compromised by inadequate sample utilization and a lack of diversity in learning strategies. To enhance MARL performance, we introduce a novel sample reuse approach that dynamically adjusts policy updates based on observation novelty. Specifically, we employ a Random Network Distillation (RND) network to gauge the novelty of each agent's current state, assigning additional sample update opportunities based on the uniqueness of the data. We name our method Multi-Agent Novelty-GuidEd sample Reuse (MANGER). This method increases sample efficiency and promotes exploration and diverse agent behaviors. Our evaluations confirm substantial improvements in MARL effectiveness in complex cooperative scenarios such as Google Research Football and super-hard StarCraft II micromanagement tasks." }, "2501.04712.pdf": { "id": "2501.04712v2", "title": "Pressing Intensity: An Intuitive Measure for Pressing in Soccer", "year": "2024", "authors": [ "Joris Bekkers" ], "categories": "stat.AP, cs.LG", "pdf_url": "https://arxiv.org/pdf/2501.04712v2", "abstract": "Pressing is a fundamental defensive strategy in football, characterized by applying pressure on the ball owning team to regain possession. Despite its significance, existing metrics for measuring pressing often lack precision or comprehensive consideration of positional data, player movement and speed. This research introduces an innovative framework for quantifying pressing intensity, leveraging advancements in positional tracking data and components from Spearman's Pitch Control model. Our method integrates player velocities, movement directions, and reaction times to compute the time required for a defender to intercept an attacker or the ball. This time-to-intercept measure is then transformed into probabilistic values using a logistic function, enabling dynamic and intuitive analysis of pressing situations at the individual frame level. the model captures how every player's movement influences pressure on the field, offering actionable insights for coaches, analysts, and decision-makers. By providing a robust and intepretable metric, our approach facilitates the identification of pressing strategies, advanced situational analyses, and the derivation of metrics, advancing the analytical capabilities for modern football." }, "2501.05870.pdf": { "id": "2501.05870v1", "title": "A Neighbor-based Approach to Pitch Ownership Models in Soccer", "year": "2025", "authors": [ "Tiago Mendes-Neves", "Lu\u00eds Meireles", "Jo\u00e3o Mendes-Moreira" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2501.05870v1", "abstract": "Pitch ownership models allow many types of analysis in soccer and provide valuable assistance to tactical analysts in understanding the game's dynamics. The novelty they provide over event-based analysis is that tracking data incorporates context that event-based data does not possess, like player positioning. This paper proposes a novel approach to building pitch ownership models in soccer games using the K-Nearest Neighbors (KNN) algorithm. Our approach provides a fast inference mechanism that can model different approaches to pitch control using the same algorithm. Despite its flexibility, it uses only three hyperparameters to tune the model, facilitating the tuning process for different player skill levels. The flexibility of the approach allows for the emulation of different methods available in the literature by adjusting a small number of parameters, including adjusting for different levels of uncertainty. In summary, the proposed model provides a new and more flexible strategy for building pitch ownership models, extending beyond just replicating existing algorithms, and can provide valuable insights for tactical analysts and open up new avenues for future research. We thoroughly visualize several examples demonstrating the presented models' strengths and weaknesses. The code is available at github.com/nvsclub/KNNPitchControl." }, "2501.05873.pdf": { "id": "2501.05873v1", "title": "Forecasting Soccer Matches through Distributions", "year": "2025", "authors": [ "Tiago Mendes-Neves", "Yassine Baghoussi", "Lu\u00eds Meireles", "Carlos Soares", "Jo\u00e3o Mendes-Moreira" ], "categories": "stat.AP", "pdf_url": "https://arxiv.org/pdf/2501.05873v1", "abstract": "Forecasting sporting events encapsulate a compelling intellectual endeavor, underscored by the substantial financial activity of an estimated $80 billion wagered in global sports betting during 2022, a trend that grows yearly. Motivated by the challenges set forth in the Springer Soccer Prediction Challenge, this study presents a method for forecasting soccer match outcomes by forecasting the shot quantity and quality distributions. The methodology integrates established ELO ratings with machine learning models. The empirical findings reveal that, despite the constraints of the challenge, this approach yields positive returns, taking advantage of the established market odds." }, "2501.09281.pdf": { "id": "2501.09281v2", "title": "SoccerSynth-Detection: A Synthetic Dataset for Soccer Player Detection", "year": "2025", "authors": [ "Haobin Qin", "Calvin Yeung", "Rikuhei Umemoto", "Keisuke Fujii" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2501.09281v2", "abstract": "In soccer video analysis, player detection is essential for identifying key events and reconstructing tactical positions. The presence of numerous players and frequent occlusions, combined with copyright restrictions, severely restricts the availability of datasets, leaving limited options such as SoccerNet-Tracking and SportsMOT. These datasets suffer from a lack of diversity, which hinders algorithms from adapting effectively to varied soccer video contexts. To address these challenges, we developed SoccerSynth-Detection, the first synthetic dataset designed for the detection of synthetic soccer players. It includes a broad range of random lighting and textures, as well as simulated camera motion blur. We validated its efficacy using the object detection model (Yolov8n) against real-world datasets (SoccerNet-Tracking and SportsMoT). In transfer tests, it matched the performance of real datasets and significantly outperformed them in images with motion blur; in pre-training tests, it demonstrated its efficacy as a pre-training dataset, significantly enhancing the algorithm's overall performance. Our work demonstrates the potential of synthetic datasets to replace real datasets for algorithm training in the field of soccer video analysis." }, "2501.18606.pdf": { "id": "2501.18606v1", "title": "Temporal dynamics of goal scoring in soccer", "year": "2025", "authors": [ "Guteraa Ayana", "Alexander Ehlert", "Joseph Ehlert", "Luca Santagata", "Maddalena Torricelli", "Brennan Klein" ], "categories": "physics.soc-ph, stat.AP", "pdf_url": "https://arxiv.org/pdf/2501.18606v1", "abstract": "We investigated the temporal distribution of goals in soccer using event-level data from 3,433 matches across 21 leagues and competitions. Contrary to the prevailing notion of randomness, we found that the probability of a goal being scored is higher as matches progress, and we observed fewer-than-expected goals in the early minutes of each half. Further analysis of the time between subsequent goals shows an exponential decay, indicating that most goals naturally cluster closer together in time. By splitting this distribution by the team that scores the next goal, we observe bursty goal-scoring dynamics, wherein the same team is more likely to score again shortly after its previous goal. These findings highlight the importance of match context (whether driven by fatigue, tactical adaptations, or psychological momentum) in shaping when teams are able to score. Moreover, the results open avenues for extending data-driven methods for identifying high-impact moments in a match and refining strategic decision-making in soccer's evolving analytical landscape." }, "2502.01613.pdf": { "id": "2502.01613v2", "title": "Statistical enhanced learning for modeling and prediction tennis matches at Grand Slam tournaments", "year": "2025", "authors": [ "Nourah Buhamra", "Andreas Groll" ], "categories": "stat.AP", "pdf_url": "https://arxiv.org/pdf/2502.01613v2", "abstract": "In this manuscript, we concentrate on a specific type of covariates, which we call statistically enhanced, for modeling tennis matches for men at Grand slam tournaments. Our goal is to assess whether these enhanced covariates have the potential to improve statistical learning approaches, in particular, with regard to the predictive performance. For this purpose, various proposed regression and machine learning model classes are compared with and without such features. To achieve this, we considered three slightly enhanced variables, namely elo rating along with two different player age variables. This concept has already been successfully applied in football, where additional team ability parameters, which were obtained from separate statistical models, were able to improve the predictive performance. In addition, different interpretable machine learning (IML) tools are employed to gain insights into the factors influencing the outcomes of tennis matches predicted by complex machine learning models, such as the random forest. Specifically, partial dependence plots (PDP) and individual conditional expectation (ICE) plots are employed to provide better interpretability for the most promising ML model from this work. Furthermore, we conduct a comparison of different regression and machine learning approaches in terms of various predictive performance measures such as classification rate, predictive Bernoulli likelihood, and Brier score. This comparison is carried out on external test data using cross-validation, rolling window, and expanding window strategies." }, "2502.02565.pdf": { "id": "2502.02565v1", "title": "Revisiting Expected Possession Value in Football: Introducing a Benchmark, U-Net Architecture, and Reward and Risk for Passes", "year": "2025", "authors": [ "Thijs Overmeer", "Tim Janssen", "Wim P. M. Nuijten" ], "categories": "cs.CV, cs.LG", "pdf_url": "https://arxiv.org/pdf/2502.02565v1", "abstract": "This paper introduces the first Expected Possession Value (EPV) benchmark and a new and improved EPV model for football. Through the introduction of the OJN-Pass-EPV benchmark, we present a novel method to quantitatively assess the quality of EPV models by using pairs of game states with given relative EPVs. Next, we attempt to replicate the results of Fern\u00e1ndez et al. (2021) using a dataset containing Dutch Eredivisie and World Cup matches. Following our failure to do so, we propose a new architecture based on U-net-type convolutional neural networks, achieving good results in model loss and Expected Calibration Error. Finally, we present an improved pass model that incorporates ball height and contains a new dual-component pass value model that analyzes reward and risk. The resulting EPV model correctly identifies the higher value state in 78% of the game state pairs in the OJN-Pass-EPV benchmark, demonstrating its ability to accurately assess goal-scoring potential. Our findings can help assess the quality of EPV models, improve EPV predictions, help assess potential reward and risk of passing decisions, and improve player and team performance." }, "2502.02785.pdf": { "id": "2502.02785v2", "title": "OpenSTARLab: Open Approach for Spatio-Temporal Agent Data Analysis in Soccer", "year": "2025", "authors": [ "Calvin Yeung", "Kenjiro Ide", "Taiga Someya", "Keisuke Fujii" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2502.02785v2", "abstract": "Sports analytics has become both more professional and sophisticated, driven by the growing availability of detailed performance data. This progress enables applications such as match outcome prediction, player scouting, and tactical analysis. In soccer, the effective utilization of event and tracking data is fundamental for capturing and analyzing the dynamics of the game. However, there are two primary challenges: the limited availability of event data, primarily restricted to top-tier teams and leagues, and the scarcity and high cost of tracking data, which complicates its integration with event data for comprehensive analysis. Here we propose OpenSTARLab, an open-source framework designed to democratize spatio-temporal agent data analysis in sports by addressing these key challenges. OpenSTARLab includes the Pre-processing Package that standardizes event and tracking data through Unified and Integrated Event Data and State-Action-Reward formats, the Event Modeling Package that implements deep learning-based event prediction, alongside the RLearn Package for reinforcement learning tasks. These technical components facilitate the handling of diverse data sources and support advanced analytical tasks, thereby enhancing the overall functionality and usability of the framework. To assess OpenSTARLab's effectiveness, we conducted several experimental evaluations. These demonstrate the superior performance of the specific event prediction model in terms of action and time prediction accuracies and maintained its robust event simulation performance. Furthermore, reinforcement learning experiments reveal a trade-off between action accuracy and temporal difference loss and show comprehensive visualization. Overall, OpenSTARLab serves as a robust platform for researchers and practitioners, enhancing innovation and collaboration in the field of soccer data analytics." }, "2502.03342.pdf": { "id": "2502.03342v1", "title": "Statistical analysis of team formation and player roles in football", "year": "2025", "authors": [ "Ali Baouan" ], "categories": "stat.AP", "pdf_url": "https://arxiv.org/pdf/2502.03342v1", "abstract": "The availability of tracking data in football presents unique opportunities for analyzing team shape and player roles, but leveraging it effectively remains challenging. This difficulty arises from the significant overlap in player positions, which complicates the identification of distinct roles and team formations. In this work, we propose a novel model that incorporates a hidden permutation matrix to simultaneously estimate team formations and assign roles to players at the frame level. To address the cardinality of permutation sets, we develop a statistical procedure to parsimoniously select relevant matrices prior to parameter estimation. Additionally, to capture formation changes during a match, we introduce a latent regime variable, enabling the modeling of dynamic tactical adjustments. This framework disentangles player locations from role-specific positions, providing a clear representation of team structure. We demonstrate the applicability of our approach using player tracking data, showcasing its potential for detailed team and player analysis." }, "2502.04864.pdf": { "id": "2502.04864v2", "title": "Redistributing Rewards Across Time and Agents for Multi-Agent Reinforcement Learning", "year": "2025", "authors": [ "Aditya Kapoor", "Kale-ab Tessera", "Mayank Baranwal", "Harshad Khadilkar", "Jan Peters", "Stefano Albrecht" ], "categories": "cs.MA, cs.AI, cs.LG, cs.RO", "pdf_url": "https://arxiv.org/pdf/2502.04864v2", "abstract": "Credit assignmen, disentangling each agent's contribution to a shared reward, is a critical challenge in cooperative multi-agent reinforcement learning (MARL). To be effective, credit assignment methods must preserve the environment's optimal policy. Some recent approaches attempt this by enforcing return equivalence, where the sum of distributed rewards must equal the team reward. However, their guarantees are conditional on a learned model's regression accuracy, making them unreliable in practice. We introduce Temporal-Agent Reward Redistribution (TAR$^2$), an approach that decouples credit modeling from this constraint. A neural network learns unnormalized contribution scores, while a separate, deterministic normalization step enforces return equivalence by construction. We demonstrate that this method is equivalent to a valid Potential-Based Reward Shaping (PBRS), which guarantees the optimal policy is preserved regardless of model accuracy. Empirically, on challenging SMACLite and Google Research Football (GRF) benchmarks, TAR$^2$ accelerates learning and achieves higher final performance than strong baselines. These results establish our method as an effective solution for the agent-temporal credit assignment problem." }, "2502.05548.pdf": { "id": "2502.05548v1", "title": "Soccer Goalkeeper Performance Evaluation: Clustering Approach", "year": "2025", "authors": [ "Mahdi Teimouri" ], "categories": "stat.AP", "pdf_url": "https://arxiv.org/pdf/2502.05548v1", "abstract": "The objective of this work was to help the soccer field managers to evaluate the performance of a goalkeeper in saving the penalty kicks. To this end, based on the concept of clustering, four measures were proposed for evaluating the goalkeeper's performance in terms of both the saved kicks and detecting the direction of kicked ball. The well-known measures ignore the goalkeeper's ability in detecting the directional jump, while the forth proposed measure in this work was regarded this fact. The effectiveness of the proposed measures were analyzed and demonstrated by evaluating the performance of goalkeepers participated in four important soccer matches. The results were consistent with the proposed measures. In summary, as well as the known point statistics for evaluating the performance of goalkeeper, the measures were proposed in this work that regard as well the ability of goalkeeper in detecting the direction of kicks are also suggested." }, "2502.07528.pdf": { "id": "2502.07528v3", "title": "Forecasting the future development in quality and value of professional football players", "year": "2025", "authors": [ "Koen W. van Arem", "Floris Goes-Smit", "Jakob S\u00f6hl" ], "categories": "stat.AP, cs.LG", "pdf_url": "https://arxiv.org/pdf/2502.07528v3", "abstract": "Transfers in professional football (soccer) are risky investments because of the large transfer fees and high risks involved. Although data-driven models can be used to improve transfer decisions, existing models focus on describing players' historical progress, leaving their future performance unknown. Moreover, recent developments have called for the use of explainable models combined with uncertainty quantification of predictions. This paper assesses explainable machine learning models based on predictive accuracy and uncertainty quantification methods for the prediction of the future development in quality and transfer value of professional football players. The predictive accuracy is studied by training the models to predict the quality and value of players one year ahead. This is carried out by training them on two data sets containing data-driven indicators describing the player quality and player value in historical settings. In general, the random forest model is found to be the most suitable model because it provides accurate predictions as well as an uncertainty quantification method that naturally arises from the bagging procedure of the random forest model. Additionally, this research shows that the development of player performance contains nonlinear patterns and interactions between variables, and that time series information can provide useful information for the modeling of player performance metrics. The resulting models can help football clubs make more informed, data-driven transfer decisions by forecasting player quality and transfer value." }, "2502.15462.pdf": { "id": "2502.15462v1", "title": "Game State and Spatio-temporal Action Detection in Soccer using Graph Neural Networks and 3D Convolutional Networks", "year": "2025", "authors": [ "Jeremie Ochin", "Guillaume Devineau", "Bogdan Stanciulescu", "Sotiris Manitsaris" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2502.15462v1", "abstract": "Soccer analytics rely on two data sources: the player positions on the pitch and the sequences of events they perform. With around 2000 ball events per game, their precise and exhaustive annotation based on a monocular video stream remains a tedious and costly manual task. While state-of-the-art spatio-temporal action detection methods show promise for automating this task, they lack contextual understanding of the game. Assuming professional players' behaviors are interdependent, we hypothesize that incorporating surrounding players' information such as positions, velocity and team membership can enhance purely visual predictions. We propose a spatio-temporal action detection approach that combines visual and game state information via Graph Neural Networks trained end-to-end with state-of-the-art 3D CNNs, demonstrating improved metrics through game state integration." }, "2502.16030.pdf": { "id": "2502.16030v1", "title": "Real Time Offside Detection using a Single Camera in Soccer", "year": "2025", "authors": [ "Shounak Desai" ], "categories": "cs.CV, cs.AI", "pdf_url": "https://arxiv.org/pdf/2502.16030v1", "abstract": "Technological advancements in soccer have surged over the past decade, transforming aspects of the sport. Unlike binary rules, many soccer regulations, such as the \"Offside Rule,\" rely on subjective interpretation rather than straightforward True or False criteria. The on-field referee holds ultimate authority in adjudicating these nuanced decisions. A significant breakthrough in soccer officiating is the Video Assistant Referee (VAR) system, leveraging a network of 20-30 cameras within stadiums to minimize human errors. VAR's operational scope typically encompasses 10-30 cameras, ensuring high decision accuracy but at a substantial cost. This report proposes an innovative approach to offside detection using a single camera, such as the broadcasting camera, to mitigate expenses associated with sophisticated technological setups." }, "2502.16496.pdf": { "id": "2502.16496v1", "title": "PMAT: Optimizing Action Generation Order in Multi-Agent Reinforcement Learning", "year": "2025", "authors": [ "Kun Hu", "Muning Wen", "Xihuai Wang", "Shao Zhang", "Yiwei Shi", "Minne Li" ], "categories": "cs.LG, cs.AI, cs.MA", "pdf_url": "https://arxiv.org/pdf/2502.16496v1", "abstract": "Multi-agent reinforcement learning (MARL) faces challenges in coordinating agents due to complex interdependencies within multi-agent systems. Most MARL algorithms use the simultaneous decision-making paradigm but ignore the action-level dependencies among agents, which reduces coordination efficiency. In contrast, the sequential decision-making paradigm provides finer-grained supervision for agent decision order, presenting the potential for handling dependencies via better decision order management. However, determining the optimal decision order remains a challenge. In this paper, we introduce Action Generation with Plackett-Luce Sampling (AGPS), a novel mechanism for agent decision order optimization. We model the order determination task as a Plackett-Luce sampling process to address issues such as ranking instability and vanishing gradient during the network training process. AGPS realizes credit-based decision order determination by establishing a bridge between the significance of agents' local observations and their decision credits, thus facilitating order optimization and dependency management. Integrating AGPS with the Multi-Agent Transformer, we propose the Prioritized Multi-Agent Transformer (PMAT), a sequential decision-making MARL algorithm with decision order optimization. Experiments on benchmarks including StarCraft II Multi-Agent Challenge, Google Research Football, and Multi-Agent MuJoCo show that PMAT outperforms state-of-the-art algorithms, greatly enhancing coordination efficiency." }, "2503.02180.pdf": { "id": "2503.02180v1", "title": "Discrete Differential Evolution Particle Swarm Optimization Algorithm for Energy Saving Flexible Job Shop Scheduling Problem Considering Machine Multi States", "year": "2025", "authors": [ "Da Wang", "Yu Zhang", "Kai Zhang", "Junqing Li", "Dengwang Li" ], "categories": "cs.NE, cs.AI", "pdf_url": "https://arxiv.org/pdf/2503.02180v1", "abstract": "As the continuous deepening of low-carbon emission reduction policies, the manufacturing industries urgently need sensible energy-saving scheduling schemes to achieve the balance between improving production efficiency and reducing energy consumption. In energy-saving scheduling, reasonable machine states-switching is a key point to achieve expected goals, i.e., whether the machines need to switch speed between different operations, and whether the machines need to add extra setup time between different jobs. Regarding this matter, this work proposes a novel machine multi states-based energy saving flexible job scheduling problem (EFJSP-M), which simultaneously takes into account machine multi speeds and setup time. To address the proposed EFJSP-M, a kind of discrete differential evolution particle swarm optimization algorithm (D-DEPSO) is designed. In specific, D-DEPSO includes a hybrid initialization strategy to improve the initial population performance, an updating mechanism embedded with differential evolution operators to enhance population diversity, and a critical path variable neighborhood search strategy to expand the solution space. At last, based on datasets DPs and MKs, the experiment results compared with five state-of-the-art algorithms demonstrate the feasible of EFJSP-M and the superior of D-DEPSO." }, "2503.09737.pdf": { "id": "2503.09737v1", "title": "Unveiling Hidden Pivotal Players with GoalNet: A GNN-Based Soccer Player Evaluation System", "year": "2025", "authors": [ "Jacky Hao Jiang", "Jerry Cai", "Anastasios Kyrillidis" ], "categories": "cs.LG, cs.AI, math.OC", "pdf_url": "https://arxiv.org/pdf/2503.09737v1", "abstract": "Soccer analysis tools emphasize metrics such as expected goals, leading to an overrepresentation of attacking players' contributions and overlooking players who facilitate ball control and link attacks. Examples include Rodri from Manchester City and Palhinha who just transferred to Bayern Munich. To address this bias, we aim to identify players with pivotal roles in a soccer team, incorporating both spatial and temporal features. In this work, we introduce a GNN-based framework that assigns individual credit for changes in expected threat (xT), thus capturing overlooked yet vital contributions in soccer. Our pipeline encodes both spatial and temporal features in event-centric graphs, enabling fair attribution of non-scoring actions such as defensive or transitional plays. We incorporate centrality measures into the learned player embeddings, ensuring that ball-retaining defenders and defensive midfielders receive due recognition for their overall impact. Furthermore, we explore diverse GNN variants-including Graph Attention Networks and Transformer-based models-to handle long-range dependencies and evolving match contexts, discussing their relative performance and computational complexity. Experiments on real match data confirm the robustness of our approach in highlighting pivotal roles that traditional attacking metrics typically miss, underscoring the model's utility for more comprehensive soccer analytics." }, "2503.11181.pdf": { "id": "2503.11181v1", "title": "Multi-Stage Generative Upscaler: Reconstructing Football Broadcast Images via Diffusion Models", "year": "2025", "authors": [ "Luca Martini", "Daniele Zolezzi", "Saverio Iacono", "Gianni Viardo Vercelli" ], "categories": "cs.CV, cs.AI", "pdf_url": "https://arxiv.org/pdf/2503.11181v1", "abstract": "The reconstruction of low-resolution football broadcast images presents a significant challenge in sports broadcasting, where detailed visuals are essential for analysis and audience engagement. This study introduces a multi-stage generative upscaling framework leveraging Diffusion Models to enhance degraded images, transforming inputs as small as $64 \\times 64$ pixels into high-fidelity $1024 \\times 1024$ outputs. By integrating an image-to-image pipeline, ControlNet conditioning, and LoRA fine-tuning, our approach surpasses traditional upscaling methods in restoring intricate textures and domain-specific elements such as player details and jersey logos. The custom LoRA is trained on a custom football dataset, ensuring adaptability to sports broadcast needs. Experimental results demonstrate substantial improvements over conventional models, with ControlNet refining fine details and LoRA enhancing task-specific elements. These findings highlight the potential of diffusion-based image reconstruction in sports media, paving the way for future applications in automated video enhancement and real-time sports analytics." }, "2503.11726.pdf": { "id": "2503.11726v1", "title": "SPECTra: Scalable Multi-Agent Reinforcement Learning with Permutation-Free Networks", "year": "2025", "authors": [ "Hyunwoo Park", "Baekryun Seong", "Sang-Ki Ko" ], "categories": "cs.LG, cs.AI", "pdf_url": "https://arxiv.org/pdf/2503.11726v1", "abstract": "In cooperative multi-agent reinforcement learning (MARL), the permutation problem where the state space grows exponentially with the number of agents reduces sample efficiency. Additionally, many existing architectures struggle with scalability, relying on a fixed structure tied to a specific number of agents, limiting their applicability to environments with a variable number of entities. While approaches such as graph neural networks (GNNs) and self-attention mechanisms have progressed in addressing these challenges, they have significant limitations as dense GNNs and self-attention mechanisms incur high computational costs. To overcome these limitations, we propose a novel agent network and a non-linear mixing network that ensure permutation-equivariance and scalability, allowing them to generalize to environments with various numbers of agents. Our agent network significantly reduces computational complexity, and our scalable hypernetwork enables efficient weight generation for non-linear mixing. Additionally, we introduce curriculum learning to improve training efficiency. Experiments on SMACv2 and Google Research Football (GRF) demonstrate that our approach achieves superior learning performance compared to existing methods. By addressing both permutation-invariance and scalability in MARL, our work provides a more efficient and adaptable framework for cooperative MARL. Our code is available at https://github.com/funny-rl/SPECTra." }, "2503.11815.pdf": { "id": "2503.11815v2", "title": "Movement Dynamics in Elite Female Soccer Athletes: The Quantile Cube Approach", "year": "2025", "authors": [ "Kendall L. Thomas", "Jan Hannig" ], "categories": "stat.ME", "pdf_url": "https://arxiv.org/pdf/2503.11815v2", "abstract": "This paper presents the quantile cube, a novel three-dimensional summary representation designed to analyze external load using GPS-derived movement data. While broadly applicable, we demonstrate its utility through an application to data from elite female soccer athletes across 23 matches. The quantile cube segments athlete movements into discrete quantiles of velocity, acceleration, and movement angle across match halves, providing a structured and interpretable framework to capture complex movement dynamics. Statistical analysis revealed significant differences in movement distributions between the first and second halves for individual athletes across all matches. Principal Component Analysis identified matches with unique movement dynamics, particularly at the start and end of the season. Dirichlet-multinomial regression further explored how factors such as athlete position, playing time, and match characteristics influenced movement profiles. Our analysis reveals external load variations over time and provides insights into performance optimization. The integration of these statistical techniques demonstrates the potential of data-driven strategies to enhance athlete monitoring and workload management in women's soccer." }, "2503.13077.pdf": { "id": "2503.13077v2", "title": "Improving Sample Efficiency in Multi-Agent Reinforcement Learning for Simulated Football Games via Exploration", "year": "2025", "authors": [ "Amir Baghi", "Jens Sj\u00f6lund", "Joakim Bergdahl", "Linus Gissl\u00e9n", "Alessandro Sestini" ], "categories": "cs.LG, cs.MA", "pdf_url": "https://arxiv.org/pdf/2503.13077v2", "abstract": "Multi-agent reinforcement learning has shown promise in learning cooperative behaviors in team-based environments. However, such methods often demand extensive training time, which inhibits their application for game-AI in standard game development. For instance, the state-of-the-art method TiZero takes 40 days to train high-quality policies for a football environment. In this paper, we hypothesize that better exploration mechanisms can improve the sample efficiency of multi-agent methods. Thereby, we propose utilizing a random network distillation bonus within the multi-agent TiZero framework, aiming to promote exploration. Additionally, we introduce architectural modifications to the original algorithm to enhance TiZero's computational efficiency. We evaluate the sample efficiency of our approach against original TiZero through extensive experiments. Our results show that random network distillation improves the sample efficiency per training phase by 13.3% compared with the original TiZero, enhancing generalization and adaptability to previously difficult scenarios. This highlights the better applicability of our variant in practical game development settings. Lastly, we qualitatively evaluate the gameplay of the produced models against a heuristic AI. We find that random network distillation leads to a higher accuracy in shooting, and it achieves higher behavioral stability as shown by the lower standard deviation achieved in gameplay metrics. The code is available at https://github.com/electronicarts/marling." }, "2503.13969.pdf": { "id": "2503.13969v2", "title": "SoccerSynth Field: enhancing field detection with synthetic data from virtual soccer simulator", "year": "2025", "authors": [ "HaoBin Qin", "Jiale Fang", "Keisuke Fujii" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2503.13969v2", "abstract": "Field detection in team sports is an essential task in sports video analysis. However, collecting large-scale and diverse real-world datasets for training detection models is often cost and time-consuming. Synthetic datasets, which allow controlled variability in lighting, textures, and camera angles, will be a promising alternative for addressing these problems. This study addresses the challenges of high costs and difficulties in collecting real-world datasets by investigating the effectiveness of pretraining models using synthetic datasets. In this paper, we propose the effectiveness of using a synthetic dataset (SoccerSynth-Field) for soccer field detection. A synthetic soccer field dataset was created to pretrain models, and the performance of these models was compared with models trained on real-world datasets. The results demonstrate that models pretrained on the synthetic dataset exhibit superior performance in detecting soccer fields. This highlights the effectiveness of synthetic data in enhancing model robustness and accuracy, offering a cost-effective and scalable solution for advancing detection tasks in sports field detection." }, "2503.14637.pdf": { "id": "2503.14637v3", "title": "KINESIS: Motion Imitation for Human Musculoskeletal Locomotion", "year": "2025", "authors": [ "Merkourios Simos", "Alberto Silvio Chiappa", "Alexander Mathis" ], "categories": "cs.RO, cs.AI, cs.CV, cs.LG, q-bio.NC", "pdf_url": "https://arxiv.org/pdf/2503.14637v3", "abstract": "How do humans move? Advances in reinforcement learning (RL) have produced impressive results in capturing human motion using physics-based humanoid control. However, torque-controlled humanoids fail to model key aspects of human motor control such as biomechanical joint constraints & non-linear and overactuated musculotendon control. We present KINESIS, a model-free motion imitation framework that tackles these challenges. KINESIS is trained on 1.8 hours of locomotion data and achieves strong motion imitation performance on unseen trajectories. Through a negative mining approach, KINESIS learns robust locomotion priors that we leverage to deploy the policy on several downstream tasks such as text-to-control, target point reaching, and football penalty kicks. Importantly, KINESIS learns to generate muscle activity patterns that correlate well with human EMG activity. We show that these results scale seamlessly across biomechanical model complexity, demonstrating control of up to 290 muscles. Overall, the physiological plausibility makes KINESIS a promising model for tackling challenging problems in human motor control. Code, videos and benchmarks are available at https://github.com/amathislab/Kinesis." }, "2503.16557.pdf": { "id": "2503.16557v1", "title": "Investigating Cultural Dimensions and Technological Acceptance: The Adoption of Electronic Performance and Tracking Systems in Qatar's Football Sector", "year": "2025", "authors": [ "Abdulaziz Al Mannai" ], "categories": "cs.CY, cs.LG", "pdf_url": "https://arxiv.org/pdf/2503.16557v1", "abstract": "Qatar's football sector has undergone a substantial technological transformation with the implementation of Electronic Performance and Tracking Systems (EPTS). This study examines the impact of cultural and technological factors on EPTS adoption, using Hofstede's Cultural Dimensions Theory and the Technology Acceptance Model (TAM) as theoretical frameworks. An initial exploratory study involved ten participants, followed by an expanded dataset comprising thirty stakeholders, including players, coaches, and staff from Qatari football organizations. Multiple regression analysis was conducted to evaluate the relationships between perceived usefulness, perceived ease of use, power distance, innovation receptiveness, integration complexity, and overall adoption. The results indicate that perceived usefulness, innovation receptiveness, and lower power distance significantly drive EPTS adoption, while ease of use is marginally significant and integration complexity is non-significant in this sample. These findings provide practical insights for sports technology stakeholders in Qatar and emphasize the importance of aligning cultural considerations with technological readiness for successful EPTS integration." }, "2503.18589.pdf": { "id": "2503.18589v2", "title": "Unified Uncertainty-Aware Diffusion for Multi-Agent Trajectory Modeling", "year": "2025", "authors": [ "Guillem Capellera", "Antonio Rubio", "Luis Ferraz", "Antonio Agudo" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2503.18589v2", "abstract": "Multi-agent trajectory modeling has primarily focused on forecasting future states, often overlooking broader tasks like trajectory completion, which are crucial for real-world applications such as correcting tracking data. Existing methods also generally predict agents' states without offering any state-wise measure of uncertainty. Moreover, popular multi-modal sampling methods lack any error probability estimates for each generated scene under the same prior observations, making it difficult to rank the predictions during inference time. We introduce U2Diff, a \\textbf{unified} diffusion model designed to handle trajectory completion while providing state-wise \\textbf{uncertainty} estimates jointly. This uncertainty estimation is achieved by augmenting the simple denoising loss with the negative log-likelihood of the predicted noise and propagating latent space uncertainty to the real state space. Additionally, we incorporate a Rank Neural Network in post-processing to enable \\textbf{error probability} estimation for each generated mode, demonstrating a strong correlation with the error relative to ground truth. Our method outperforms the state-of-the-art solutions in trajectory completion and forecasting across four challenging sports datasets (NBA, Basketball-U, Football-U, Soccer-U), highlighting the effectiveness of uncertainty and error probability estimation. Video at https://youtu.be/ngw4D4eJToE" }, "2503.19809.pdf": { "id": "2503.19809v1", "title": "Simulating Tracking Data to Advance Sports Analytics Research", "year": "2025", "authors": [ "David Radke", "Kyle Tilbury" ], "categories": "cs.AI", "pdf_url": "https://arxiv.org/pdf/2503.19809v1", "abstract": "Advanced analytics have transformed how sports teams operate, particularly in episodic sports like baseball. Their impact on continuous invasion sports, such as soccer and ice hockey, has been limited due to increased game complexity and restricted access to high-resolution game tracking data. In this demo, we present a method to collect and utilize simulated soccer tracking data from the Google Research Football environment to support the development of models designed for continuous tracking data. The data is stored in a schema that is representative of real tracking data and we provide processes that extract high-level features and events. We include examples of established tracking data models to showcase the efficacy of the simulated data. We address the scarcity of publicly available tracking data, providing support for research at the intersection of artificial intelligence and sports analytics." }, "2504.00767.pdf": { "id": "2504.00767v1", "title": "Automated Explanation of Machine Learning Models of Footballing Actions in Words", "year": "2025", "authors": [ "Pegah Rahimian", "Jernej Flisar", "David Sumpter" ], "categories": "cs.LG, cs.CL, cs.HC", "pdf_url": "https://arxiv.org/pdf/2504.00767v1", "abstract": "While football analytics has changed the way teams and analysts assess performance, there remains a communication gap between machine learning practice and how coaching staff talk about football. Coaches and practitioners require actionable insights, which are not always provided by models. To bridge this gap, we show how to build wordalizations (a novel approach that leverages large language models) for shots in football. Specifically, we first build an expected goals model using logistic regression. We then use the co-efficients of this regression model to write sentences describing how factors (such as distance, angle and defensive pressure) contribute to the model's prediction. Finally, we use large language models to give an entertaining description of the shot. We describe our approach in a model card and provide an interactive open-source application describing shots in recent tournaments. We discuss how shot wordalisations might aid communication in coaching and football commentary, and give a further example of how the same approach can be applied to other actions in football." }, "2504.02184.pdf": { "id": "2504.02184v2", "title": "Model Predictive Control with Visibility Graphs for Humanoid Path Planning and Tracking Against Adversarial Opponents", "year": "2025", "authors": [ "Ruochen Hou", "Gabriel I. Fernandez", "Mingzhang Zhu", "Dennis W. Hong" ], "categories": "cs.RO, eess.SY", "pdf_url": "https://arxiv.org/pdf/2504.02184v2", "abstract": "In this paper we detail the methods used for obstacle avoidance, path planning, and trajectory tracking that helped us win the adult-sized, autonomous humanoid soccer league in RoboCup 2024. Our team was undefeated for all seated matches and scored 45 goals over 6 games, winning the championship game 6 to 1. During the competition, a major challenge for collision avoidance was the measurement noise coming from bipedal locomotion and a limited field of view (FOV). Furthermore, obstacles would sporadically jump in and out of our planned trajectory. At times our estimator would place our robot inside a hard constraint. Any planner in this competition must also be be computationally efficient enough to re-plan and react in real time. This motivated our approach to trajectory generation and tracking. In many scenarios long-term and short-term planning is needed. To efficiently find a long-term general path that avoids all obstacles we developed DAVG (Dynamic Augmented Visibility Graphs). DAVG focuses on essential path planning by setting certain regions to be active based on obstacles and the desired goal pose. By augmenting the states in the graph, turning angles are considered, which is crucial for a large soccer playing robot as turning may be more costly. A trajectory is formed by linearly interpolating between discrete points generated by DAVG. A modified version of model predictive control (MPC) is used to then track this trajectory called cf-MPC (Collision-Free MPC). This ensures short-term planning. Without having to switch formulations cf-MPC takes into account the robot dynamics and collision free constraints. Without a hard switch the control input can smoothly transition in cases where the noise places our robot inside a constraint boundary. The nonlinear formulation runs at approximately 120 Hz, while the quadratic version achieves around 400 Hz." }, "2504.06357.pdf": { "id": "2504.06357v1", "title": "From Broadcast to Minimap: Achieving State-of-the-Art SoccerNet Game State Reconstruction", "year": "2025", "authors": [ "Vladimir Golovkin", "Nikolay Nemtsev", "Vasyl Shandyba", "Oleg Udin", "Nikita Kasatkin", "Pavel Kononov" ], "categories": "cs.CV, cs.LG", "pdf_url": "https://arxiv.org/pdf/2504.06357v1", "abstract": "Game State Reconstruction (GSR), a critical task in Sports Video Understanding, involves precise tracking and localization of all individuals on the football field-players, goalkeepers, referees, and others - in real-world coordinates. This capability enables coaches and analysts to derive actionable insights into player movements, team formations, and game dynamics, ultimately optimizing training strategies and enhancing competitive advantage. Achieving accurate GSR using a single-camera setup is highly challenging due to frequent camera movements, occlusions, and dynamic scene content. In this work, we present a robust end-to-end pipeline for tracking players across an entire match using a single-camera setup. Our solution integrates a fine-tuned YOLOv5m for object detection, a SegFormer-based camera parameter estimator, and a DeepSORT-based tracking framework enhanced with re-identification, orientation prediction, and jersey number recognition. By ensuring both spatial accuracy and temporal consistency, our method delivers state-of-the-art game state reconstruction, securing first place in the SoccerNet Game State Reconstruction Challenge 2024 and significantly outperforming competing methods." }, "2504.09499.pdf": { "id": "2504.09499v2", "title": "Decoding the mechanisms of the Hattrick football manager game using Bayesian network structure learning", "year": "2025", "authors": [ "Anthony C. Constantinou", "Nicholas Higgins", "Neville K. Kitson" ], "categories": "cs.LG, cs.AI", "pdf_url": "https://arxiv.org/pdf/2504.09499v2", "abstract": "Hattrick is a free web-based probabilistic football manager game with over 200,000 users competing for titles at national and international levels. Launched in Sweden in 1997 as part of an MSc project, the game's slow-paced design has fostered a loyal community, with users remaining active for decades. Hattrick's game-engine mechanics are partially hidden, and users have attempted to decode them with incremental success over the years. Rule-based, statistical and machine learning models have been developed to aid this effort and are widely used by the community, but have not been formally evaluated in the scientific literature. This study is the first to explore Hattrick using structure learning techniques and Bayesian networks, integrating expert knowledge with data to develop models that simulate and explain the game-engine. We assess the effectiveness of structure learning algorithms in relation to knowledge-based structures, and publicly share a fully specified Bayesian network model that matches the performance of top models used by the Hattrick community. We further demonstrate how analysis extends beyond prediction by providing a visual representation of dependencies between features, and using the optimal model for in-game decision-making. To support future research, we make all data, graphical structures, and models publicly available online." }, "2504.12021.pdf": { "id": "2504.12021v1", "title": "Action Anticipation from SoccerNet Football Video Broadcasts", "year": "2025", "authors": [ "Mohamad Dalal", "Artur Xarles", "Anthony Cioppa", "Silvio Giancola", "Marc Van Droogenbroeck", "Bernard Ghanem" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2504.12021v1", "abstract": "Artificial intelligence has revolutionized the way we analyze sports videos, whether to understand the actions of games in long untrimmed videos or to anticipate the player's motion in future frames. Despite these efforts, little attention has been given to anticipating game actions before they occur. In this work, we introduce the task of action anticipation for football broadcast videos, which consists in predicting future actions in unobserved future frames, within a five- or ten-second anticipation window. To benchmark this task, we release a new dataset, namely the SoccerNet Ball Action Anticipation dataset, based on SoccerNet Ball Action Spotting. Additionally, we propose a Football Action ANticipation TRAnsformer (FAANTRA), a baseline method that adapts FUTR, a state-of-the-art action anticipation model, to predict ball-related actions. To evaluate action anticipation, we introduce new metrics, including mAP@$\u03b4$, which evaluates the temporal precision of predicted future actions, as well as mAP@$\\infty$, which evaluates their occurrence within the anticipation window. We also conduct extensive ablation studies to examine the impact of various task settings, input configurations, and model architectures. Experimental results highlight both the feasibility and challenges of action anticipation in football videos, providing valuable insights into the design of predictive models for sports analytics. By forecasting actions before they unfold, our work will enable applications in automated broadcasting, tactical analysis, and player decision-making. Our dataset and code are publicly available at https://github.com/MohamadDalal/FAANTRA." }, "2504.13286.pdf": { "id": "2504.13286v1", "title": "A Model Predictive Control Approach for Quadrotor Cruise Control", "year": "2025", "authors": [ "Zekai Chen", "Leon Kehler" ], "categories": "eess.SY", "pdf_url": "https://arxiv.org/pdf/2504.13286v1", "abstract": "This paper investigates the application of a Model Predictive Controller (MPC) for the cruise control system of a quadrotor, focusing on hovering point stabilization and reference tracking. Initially, a full-state-feedback MPC is designed for the ideal scenario. To account for real-world conditions, a constant disturbance is introduced to the quadrotor, simulating a gust of wind in a specific direction. In response, an output-feedback offset-free MPC is developed to stabilize the quadrotor while rejecting the disturbance. We validate the design of the controller by conducting stability analysis, as well as numerical simulations under different circumstances. It is shown that the designed controller can achieve all the expected goals for the cruise control, including reference tracking and disturbance rejection. This project was implemented using Python and the CVXPY library for convex optimization." }, "2504.16572.pdf": { "id": "2504.16572v1", "title": "Bridging Data Gaps and Building Knowledge Networks in Indian Football Analytics", "year": "2025", "authors": [ "Sneha Nanavati", "Nimmi Rangaswamy" ], "categories": "cs.HC", "pdf_url": "https://arxiv.org/pdf/2504.16572v1", "abstract": "The global rise of football analytics has rapidly transformed how clubs make strategic decisions. However, in India, the adoption of analytics remains constrained by institutional resistance, infrastructural limitations, and cultural barriers -- challenges that grassroots innovation and low-cost data solutions have the potential to overcome. Despite the growing popularity of the Indian Super League, resource scarcity and fragmented governance continue to hinder the widespread adoption and impact of analytics. This mixed-methods study explores how informal, decentralised analytics communities -- comprising amateur analysts and Twitter-based \"data sleuths\" -- navigate these constraints through peer mentorship and grassroots innovation. Drawing on extensive digital ethnography, participant observation, and interviews, the study illustrates how these informal networks mitigate data scarcity, limited digital infrastructure, and institutional indifference while fostering skill development and professional growth. Building on these insights, the paper proposes HCI interventions such as decentralised knowledge platforms to facilitate structured, cross-border peer mentorship and low-cost data solutions -- including AI-assisted player tracking and mobile analytics dashboards -- rooted in principles of frugal innovation. These interventions aim to bridge the data divide, support inclusive technical engagement in sport, and enhance analytics-driven decision-making in resource-constrained environments. This paper contributes to HCIxB's focus on cross-border collaboration by highlighting how community-driven technological adaptation in the Global South can foster meaningful participation, skill-building, and long-term sustainability through informal learning networks and scalable, context-sensitive tools." }, "2504.17365.pdf": { "id": "2504.17365v3", "title": "TimeSoccer: An End-to-End Multimodal Large Language Model for Soccer Commentary Generation", "year": "2025", "authors": [ "Ling You", "Wenxuan Huang", "Xinni Xie", "Xiangyi Wei", "Bangyan Li", "Shaohui Lin" ], "categories": "cs.CV, cs.CL", "pdf_url": "https://arxiv.org/pdf/2504.17365v3", "abstract": "Soccer is a globally popular sporting event, typically characterized by long matches and distinctive highlight moments. Recent advances in Multimodal Large Language Models (MLLMs) offer promising capabilities in temporal grounding and video understanding, soccer commentary generation often requires precise temporal localization and semantically rich descriptions over long-form video. However, existing soccer MLLMs often rely on the temporal a priori for caption generation, so they cannot process the soccer video end-to-end. While some traditional approaches follow a two-step paradigm that is complex and fails to capture the global context to achieve suboptimal performance. To solve the above issues, we present TimeSoccer, the first end-to-end soccer MLLM for Single-anchor Dense Video Captioning (SDVC) in full-match soccer videos. TimeSoccer jointly predicts timestamps and generates captions in a single pass, enabling global context modeling across 45-minute matches. To support long video understanding of soccer matches, we introduce MoFA-Select, a training-free, motion-aware frame compression module that adaptively selects representative frames via a coarse-to-fine strategy, and incorporates complementary training paradigms to strengthen the model's ability to handle long temporal sequences. Extensive experiments demonstrate that our TimeSoccer achieves State-of-The-Art (SoTA) performance on the SDVC task in an end-to-end form, generating high-quality commentary with accurate temporal alignment and strong semantic relevance." }, "2504.19612.pdf": { "id": "2504.19612v2", "title": "When Should Team KPIs Be Absolute or Relative for Match-Outcome Prediction?", "year": "2025", "authors": [ "M. R. Brown", "G. Scott", "L. Kilduff" ], "categories": "physics.data-an", "pdf_url": "https://arxiv.org/pdf/2504.19612v2", "abstract": "In rugby union and association football, team key performance indicators (KPIs) can be represented in absolute terms or relative to the opponent. Relativisation sometimes improves match-outcome prediction and sometimes harms it. There has been no general account of when each occurs. The answer depends on how much the two teams differ in variability and how strongly their KPI values rise and fall together. We combine these properties into the Paired Efficiency Factor (PEF), which generalises Fisher's paired-efficiency result to the unequal-variance conditions typical of competitive sport. The PEF also connects a KPI's statistical efficiency to how much information its relative form carries about the outcome. Combined KPIs can interact in complex ways, so we analyse each indicator on its own. Across 86 team KPIs from professional rugby union and association football, the PEF places every metric in one of four regimes. Anti-correlation is common, especially for high-volume competitive counts, and absolute measures are then usually preferable. Relativisation can still improve prediction when a noisier difference carries more outcome information. An idealised simulation and one representative KPI from each regime confirm both signs under team-blocked cross-validation. The same pairing geometry appears in healthcare, genomics, finance, and manufacturing. The PEF turns an ad hoc feature-engineering choice into a transparent, data-informed diagnostic for when to relativise performance metrics and when not to." }, "2504.20808.pdf": { "id": "2504.20808v2", "title": "SoccerDiffusion: Toward Learning End-to-End Humanoid Robot Soccer from Gameplay Recordings", "year": "2025", "authors": [ "Florian Vahl", "J\u00f6rn Griepenburg", "Jan Gutsche", "Jasper G\u00fcldenstein", "Jianwei Zhang" ], "categories": "cs.RO, cs.AI, cs.LG", "pdf_url": "https://arxiv.org/pdf/2504.20808v2", "abstract": "This paper introduces SoccerDiffusion, a transformer-based diffusion model designed to learn end-to-end control policies for humanoid robot soccer directly from real-world gameplay recordings. Using data collected from RoboCup competitions, the model predicts joint command trajectories from multi-modal sensor inputs, including vision, proprioception, and game state. We employ a distillation technique to enable real-time inference on embedded platforms that reduces the multi-step diffusion process to a single step. Our results demonstrate the model's ability to replicate complex motion behaviors such as walking, kicking, and fall recovery both in simulation and on physical robots. Although high-level tactical behavior remains limited, this work provides a robust foundation for subsequent reinforcement learning or preference optimization methods. We release the dataset, pretrained models, and code under: https://bit-bots.github.io/SoccerDiffusion" }, "2505.03735.pdf": { "id": "2505.03735v2", "title": "Multi-Agent System for Comprehensive Soccer Understanding", "year": "2025", "authors": [ "Jiayuan Rao", "Zifeng Li", "Haoning Wu", "Ya Zhang", "Yanfeng Wang", "Weidi Xie" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2505.03735v2", "abstract": "Recent advances in soccer understanding have demonstrated rapid progress, yet existing research predominantly focuses on isolated or narrow tasks. To bridge this gap, we propose a comprehensive framework for holistic soccer understanding. Concretely, we make the following contributions in this paper: (i) we construct SoccerWiki, the first large-scale multimodal soccer knowledge base, integrating rich domain knowledge about players, teams, referees, and venues to enable knowledge-driven reasoning; (ii) we present SoccerBench, the largest and most comprehensive soccer-specific benchmark, featuring around 10K multimodal (text, image, video) multi-choice QA pairs across 13 distinct tasks; (iii) we introduce SoccerAgent, a novel multi-agent system that decomposes complex soccer questions via collaborative reasoning, leveraging domain expertise from SoccerWiki and achieving robust performance; (iv) extensive evaluations and comparisons with representative MLLMs on SoccerBench highlight the superiority of our agentic system." }, "2505.09455.pdf": { "id": "2505.09455v1", "title": "Beyond Pixels: Leveraging the Language of Soccer to Improve Spatio-Temporal Action Detection in Broadcast Videos", "year": "2025", "authors": [ "Jeremie Ochin", "Raphael Chekroun", "Bogdan Stanciulescu", "Sotiris Manitsaris" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2505.09455v1", "abstract": "State-of-the-art spatio-temporal action detection (STAD) methods show promising results for extracting soccer events from broadcast videos. However, when operated in the high-recall, low-precision regime required for exhaustive event coverage in soccer analytics, their lack of contextual understanding becomes apparent: many false positives could be resolved by considering a broader sequence of actions and game-state information. In this work, we address this limitation by reasoning at the game level and improving STAD through the addition of a denoising sequence transduction task. Sequences of noisy, context-free player-centric predictions are processed alongside clean game state information using a Transformer-based encoder-decoder model. By modeling extended temporal context and reasoning jointly over team-level dynamics, our method leverages the \"language of soccer\" - its tactical regularities and inter-player dependencies - to generate \"denoised\" sequences of actions. This approach improves both precision and recall in low-confidence regimes, enabling more reliable event extraction from broadcast video and complementing existing pixel-based methods." }, "2505.13834.pdf": { "id": "2505.13834v2", "title": "Toward Real-World Cooperative and Competitive Soccer with Quadrupedal Robot Teams", "year": "2025", "authors": [ "Zhi Su", "Yuman Gao", "Emily Lukas", "Yunfei Li", "Jiaze Cai", "Faris Tulbah" ], "categories": "cs.RO, cs.AI", "pdf_url": "https://arxiv.org/pdf/2505.13834v2", "abstract": "Achieving coordinated teamwork among legged robots requires both fine-grained locomotion control and long-horizon strategic decision-making. Robot soccer offers a compelling testbed for this challenge, combining dynamic, competitive, and multi-agent interactions. In this work, we present a hierarchical multi-agent reinforcement learning (MARL) framework that enables fully autonomous and decentralized quadruped robot soccer. First, a set of highly dynamic low-level skills is trained for legged locomotion and ball manipulation, such as walking, dribbling, and kicking. On top of these, a high-level strategic planning policy is trained with Multi-Agent Proximal Policy Optimization (MAPPO) via Fictitious Self-Play (FSP). This learning framework allows agents to adapt to diverse opponent strategies and gives rise to sophisticated team behaviors, including coordinated passing, interception, and dynamic role allocation. With an extensive ablation study, the proposed learning method shows significant advantages in the cooperative and competitive multi-agent soccer game. We deploy the learned policies to real quadruped robots relying solely on onboard proprioception and decentralized localization, with the resulting system supporting autonomous robot-robot and robot-human soccer matches on indoor and outdoor soccer courts." }, "2505.13860.pdf": { "id": "2505.13860v2", "title": "Domain Adaptation of VLM for Soccer Video Understanding", "year": "2025", "authors": [ "Tiancheng Jiang", "Henry Wang", "Md Sirajus Salekin", "Parmida Atighehchian", "Shinan Zhang" ], "categories": "cs.CV, cs.AI", "pdf_url": "https://arxiv.org/pdf/2505.13860v2", "abstract": "Vision Language Models (VLMs) have demonstrated strong performance in multi-modal tasks by effectively aligning visual and textual representations. However, most video understanding VLM research has been domain-agnostic, leaving the understanding of their transfer learning capability to specialized domains under-explored. In this work, we address this by exploring the adaptability of open-source VLMs to specific domains, and focusing on soccer as an initial case study. Our approach uses large-scale soccer datasets and LLM to create instruction-following data, and use them to iteratively fine-tune the general-domain VLM in a curriculum learning fashion (first teaching the model key soccer concepts to then question answering tasks). The final adapted model, trained using a curated dataset of 20k video clips, exhibits significant improvement in soccer-specific tasks compared to the base model, with a 37.5% relative improvement for the visual question-answering task and an accuracy improvement from 11.8% to 63.5% for the downstream soccer action classification task." }, "2505.16630.pdf": { "id": "2505.16630v1", "title": "SoccerChat: Integrating Multimodal Data for Enhanced Soccer Game Understanding", "year": "2025", "authors": [ "Sushant Gautam", "Cise Midoglu", "Vajira Thambawita", "Michael A. Riegler", "P\u00e5l Halvorsen", "Mubarak Shah" ], "categories": "cs.CV, cs.AI", "pdf_url": "https://arxiv.org/pdf/2505.16630v1", "abstract": "The integration of artificial intelligence in sports analytics has transformed soccer video understanding, enabling real-time, automated insights into complex game dynamics. Traditional approaches rely on isolated data streams, limiting their effectiveness in capturing the full context of a match. To address this, we introduce SoccerChat, a multimodal conversational AI framework that integrates visual and textual data for enhanced soccer video comprehension. Leveraging the extensive SoccerNet dataset, enriched with jersey color annotations and automatic speech recognition (ASR) transcripts, SoccerChat is fine-tuned on a structured video instruction dataset to facilitate accurate game understanding, event classification, and referee decision making. We benchmark SoccerChat on action classification and referee decision-making tasks, demonstrating its performance in general soccer event comprehension while maintaining competitive accuracy in referee decision making. Our findings highlight the importance of multimodal integration in advancing soccer analytics, paving the way for more interactive and explainable AI-driven sports analysis. https://github.com/simula/SoccerChat" }, "2506.04251.pdf": { "id": "2506.04251v4", "title": "Language-Driven Coordination and Learning in Multi-Agent Simulation Environments", "year": "2025", "authors": [ "Zhengyang Li", "Sawyer Campos", "Nana Wang" ], "categories": "cs.AI, cs.LG, cs.MA", "pdf_url": "https://arxiv.org/pdf/2506.04251v4", "abstract": "This paper introduces LLM-MARL, a unified framework that incorporates large language models (LLMs) into multi-agent reinforcement learning (MARL) to enhance coordination, communication, and generalization in simulated game environments. The framework features three modular components of Coordinator, Communicator, and Memory, which dynamically generate subgoals, facilitate symbolic inter-agent messaging, and support episodic recall. Training combines PPO with a language-conditioned loss and LLM query gating. LLM-MARL is evaluated in Google Research Football, MAgent Battle, and StarCraft II. Results show consistent improvements over MAPPO and QMIX in win rate, coordination score, and zero-shot generalization. Ablation studies demonstrate that subgoal generation and language-based messaging each contribute significantly to performance gains. Qualitative analysis reveals emergent behaviors such as role specialization and communication-driven tactics. By bridging language modeling and policy learning, this work contributes to the design of intelligent, cooperative agents in interactive simulations. It offers a path forward for leveraging LLMs in multi-agent systems used for training, games, and human-AI collaboration." }, "2506.06666.pdf": { "id": "2506.06666v1", "title": "Through the Gaps: Uncovering Tactical Line-Breaking Passes with Clustering", "year": "2025", "authors": [ "Oktay Karaku\u015f", "Hasan Arkada\u015f" ], "categories": "cs.LG, stat.ML", "pdf_url": "https://arxiv.org/pdf/2506.06666v1", "abstract": "Line-breaking passes (LBPs) are crucial tactical actions in football, allowing teams to penetrate defensive lines and access high-value spaces. In this study, we present an unsupervised, clustering-based framework for detecting and analysing LBPs using synchronised event and tracking data from elite matches. Our approach models opponent team shape through vertical spatial segmentation and identifies passes that disrupt defensive lines within open play. Beyond detection, we introduce several tactical metrics, including the space build-up ratio (SBR) and two chain-based variants, LBPCh$^1$ and LBPCh$^2$, which quantify the effectiveness of LBPs in generating immediate or sustained attacking threats. We evaluate these metrics across teams and players in the 2022 FIFA World Cup, revealing stylistic differences in vertical progression and structural disruption. The proposed methodology is explainable, scalable, and directly applicable to modern performance analysis and scouting workflows." }, "2506.07981.pdf": { "id": "2506.07981v1", "title": "Real-time Localization of a Soccer Ball from a Single Camera", "year": "2025", "authors": [ "Dmitrii Vorobev", "Artem Prosvetov", "Karim Elhadji Daou" ], "categories": "cs.CV, cs.LG", "pdf_url": "https://arxiv.org/pdf/2506.07981v1", "abstract": "We propose a computationally efficient method for real-time three-dimensional football trajectory reconstruction from a single broadcast camera. In contrast to previous work, our approach introduces a multi-mode state model with $W$ discrete modes to significantly accelerate optimization while preserving centimeter-level accuracy -- even in cases of severe occlusion, motion blur, and complex backgrounds. The system operates on standard CPUs and achieves low latency suitable for live broadcast settings. Extensive evaluation on a proprietary dataset of 6K-resolution Russian Premier League matches demonstrates performance comparable to multi-camera systems, without the need for specialized or costly infrastructure. This work provides a practical method for accessible and accurate 3D ball tracking in professional football environments." }, "2506.11399.pdf": { "id": "2506.11399v1", "title": "Time-Varying Home Field Advantage in Football: Learning from a Non-Stationary Causal Process", "year": "2025", "authors": [ "Minhao Qi", "Hengrui Cai", "Guanyu Hu", "Weining Shen" ], "categories": "stat.AP, stat.ML", "pdf_url": "https://arxiv.org/pdf/2506.11399v1", "abstract": "In sports analytics, home field advantage is a robust phenomenon where the home team wins more games than the away team. However, discovering the causal factors behind home field advantage presents unique challenges due to the non-stationary, time-varying environment of sports matches. In response, we propose a novel causal discovery method, DYnamic Non-stAtionary local M-estimatOrs (DYNAMO), to learn the time-varying causal structures of home field advantage. DYNAMO offers flexibility by integrating various loss functions, making it practical for learning linear and non-linear causal structures from a general class of non-stationary causal processes. By leveraging local information, we provide theoretical guarantees for the identifiability and estimation consistency of non-stationary causal structures without imposing additional assumptions. Simulation studies validate the efficacy of DYNAMO in recovering time-varying causal structures. We apply our method to high-resolution event data from the 2020-2021 and 2021-2022 English Premier League seasons, during which the former season had no audience presence. Our results reveal intriguing, time-varying, team-specific field advantages influenced by referee bias, which differ significantly with and without crowd support. Furthermore, the time-varying causal structures learned by our method improve goal prediction accuracy compared to existing methods." }, "2506.17144.pdf": { "id": "2506.17144v2", "title": "Do We Need Large VLMs for Spotting Soccer Actions?", "year": "2025", "authors": [ "Ritabrata Chakraborty", "Rajatsubhra Chakraborty", "Avijit Dasgupta", "Sandeep Chaurasia" ], "categories": "cs.CV, cs.AI, cs.LG", "pdf_url": "https://arxiv.org/pdf/2506.17144v2", "abstract": "Traditional video-based tasks like soccer action spotting rely heavily on visual inputs, often requiring complex and computationally expensive models to process dense video data. We propose a shift from this video-centric approach to a text-based task, making it lightweight and scalable by utilizing Large Language Models (LLMs) instead of Vision-Language Models (VLMs). We posit that expert commentary, which provides rich descriptions and contextual cues contains sufficient information to reliably spot key actions in a match. To demonstrate this, we employ a system of three LLMs acting as judges specializing in outcome, excitement, and tactics for spotting actions in soccer matches. Our experiments show that this language-centric approach performs effectively in detecting critical match events coming close to state-of-the-art video-based spotters while using zero video processing compute and similar amount of time to process the entire match." }, "2506.21784.pdf": { "id": "2506.21784v1", "title": "MobiVerse: Scaling Urban Mobility Simulation with Hybrid Lightweight Domain-Specific Generator and Large Language Models", "year": "2025", "authors": [ "Yifan Liu", "Xishun Liao", "Haoxuan Ma", "Jonathan Liu", "Rohan Jadhav", "Jiaqi Ma" ], "categories": "cs.AI", "pdf_url": "https://arxiv.org/pdf/2506.21784v1", "abstract": "Understanding and modeling human mobility patterns is crucial for effective transportation planning and urban development. Despite significant advances in mobility research, there remains a critical gap in simulation platforms that allow for algorithm development, policy implementation, and comprehensive evaluation at scale. Traditional activity-based models require extensive data collection and manual calibration, machine learning approaches struggle with adaptation to dynamic conditions, and treding agent-based Large Language Models (LLMs) implementations face computational constraints with large-scale simulations. To address these challenges, we propose MobiVerse, a hybrid framework leverages the efficiency of lightweight domain-specific generator for generating base activity chains with the adaptability of LLMs for context-aware modifications. A case study was conducted in Westwood, Los Angeles, where we efficiently generated and dynamically adjusted schedules for the whole population of approximately 53,000 agents on a standard PC. Our experiments demonstrate that MobiVerse successfully enables agents to respond to environmental feedback, including road closures, large gathering events like football games, and congestion, through our hybrid framework. Its modular design facilitates testing various mobility algorithms at both transportation system and agent levels. Results show our approach maintains computational efficiency while enhancing behavioral realism. MobiVerse bridges the gap in mobility simulation by providing a customizable platform for mobility systems planning and operations with benchmark algorithms. Code and videos are available at https://github.com/ucla-mobility/MobiVerse." }, "2506.22503.pdf": { "id": "2506.22503v1", "title": "What Makes a Dribble Successful? Insights From 3D Pose Tracking Data", "year": "2025", "authors": [ "Michiel Schepers", "Pieter Robberechts", "Jan Van Haaren", "Jesse Davis" ], "categories": "cs.CV, cs.LG", "pdf_url": "https://arxiv.org/pdf/2506.22503v1", "abstract": "Data analysis plays an increasingly important role in soccer, offering new ways to evaluate individual and team performance. One specific application is the evaluation of dribbles: one-on-one situations where an attacker attempts to bypass a defender with the ball. While previous research has primarily relied on 2D positional tracking data, this fails to capture aspects like balance, orientation, and ball control, limiting the depth of current insights. This study explores how pose tracking data (capturing players' posture and movement in three dimensions) can improve our understanding of dribbling skills. We extract novel pose-based features from 1,736 dribbles in the 2022/23 Champions League season and evaluate their impact on dribble success. Our results indicate that features capturing the attacker's balance and the alignment of the orientation between the attacker and defender are informative for predicting dribble success. Incorporating these pose-based features on top of features derived from traditional 2D positional data leads to a measurable improvement in model performance." }, "2506.23843.pdf": { "id": "2506.23843v1", "title": "EFPI: Elastic Formation and Position Identification in Football (Soccer) using Template Matching and Linear Assignment", "year": "2025", "authors": [ "Joris Bekkers" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2506.23843v1", "abstract": "Understanding team formations and player positioning is crucial for tactical analysis in football (soccer). This paper presents a flexible method for formation recognition and player position assignment in football using predefined static formation templates and cost minimization from spatiotemporal tracking data, called EFPI. Our approach employs linear sum assignment to optimally match players to positions within a set of template formations by minimizing the total distance between actual player locations and template positions, subsequently selecting the formation with the lowest assignment cost. To improve accuracy, we scale actual player positions to match the dimensions of these formation templates in both width and length. While the method functions effectively on individual frames, it extends naturally to larger game segments such as complete periods, possession sequences or specific intervals (e.g. 10 second intervals, 5 minute intervals etc.). Additionally, we incorporate an optional stability parameter that prevents unnecessary formation changes when assignment costs differ only marginally between time segments. EFPI is available as open-source code through the unravelsports Python package." }, "2507.10626.pdf": { "id": "2507.10626v1", "title": "Player-Team Heterogeneous Interaction Graph Transformer for Soccer Outcome Prediction", "year": "2025", "authors": [ "Lintao Wang", "Shiwen Xu", "Michael Horton", "Joachim Gudmundsson", "Zhiyong Wang" ], "categories": "cs.LG, cs.AI", "pdf_url": "https://arxiv.org/pdf/2507.10626v1", "abstract": "Predicting soccer match outcomes is a challenging task due to the inherently unpredictable nature of the game and the numerous dynamic factors influencing results. While it conventionally relies on meticulous feature engineering, deep learning techniques have recently shown a great promise in learning effective player and team representations directly for soccer outcome prediction. However, existing methods often overlook the heterogeneous nature of interactions among players and teams, which is crucial for accurately modeling match dynamics. To address this gap, we propose HIGFormer (Heterogeneous Interaction Graph Transformer), a novel graph-augmented transformer-based deep learning model for soccer outcome prediction. HIGFormer introduces a multi-level interaction framework that captures both fine-grained player dynamics and high-level team interactions. Specifically, it comprises (1) a Player Interaction Network, which encodes player performance through heterogeneous interaction graphs, combining local graph convolutions with a global graph-augmented transformer; (2) a Team Interaction Network, which constructs interaction graphs from a team-to-team perspective to model historical match relationships; and (3) a Match Comparison Transformer, which jointly analyzes both team and player-level information to predict match outcomes. Extensive experiments on the WyScout Open Access Dataset, a large-scale real-world soccer dataset, demonstrate that HIGFormer significantly outperforms existing methods in prediction accuracy. Furthermore, we provide valuable insights into leveraging our model for player performance evaluation, offering a new perspective on talent scouting and team strategy analysis." }, "2507.12504.pdf": { "id": "2507.12504v1", "title": "Transforming Football Data into Object-centric Event Logs with Spatial Context Information", "year": "2025", "authors": [ "Vito Chan", "Lennart Ebert", "Paul-Julius Hillmann", "Christoffer Rubensson", "Stephan A. Fahrenkrog-Petersen", "Jan Mendling" ], "categories": "cs.DB, cs.AI", "pdf_url": "https://arxiv.org/pdf/2507.12504v1", "abstract": "Object-centric event logs expand the conventional single-case notion event log by considering multiple objects, allowing for the analysis of more complex and realistic process behavior. However, the number of real-world object-centric event logs remains limited, and further studies are needed to test their usefulness. The increasing availability of data from team sports can facilitate object-centric process mining, leveraging both real-world data and suitable use cases. In this paper, we present a framework for transforming football (soccer) data into an object-centric event log, further enhanced with a spatial dimension. We demonstrate the effectiveness of our framework by generating object-centric event logs based on real-world football data and discuss the results for varying process representations. With our paper, we provide the first example for object-centric event logs in football analytics. Future work should consider variant analysis and filtering techniques to better handle variability" }, "2507.12617.pdf": { "id": "2507.12617v1", "title": "Predicting Soccer Penalty Kick Direction Using Human Action Recognition", "year": "2025", "authors": [ "David Freire-Obreg\u00f3n", "Oliverio J. Santana", "Javier Lorenzo-Navarro", "Daniel Hern\u00e1ndez-Sosa", "Modesto Castrill\u00f3n-Santana" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2507.12617v1", "abstract": "Action anticipation has become a prominent topic in Human Action Recognition (HAR). However, its application to real-world sports scenarios remains limited by the availability of suitable annotated datasets. This work presents a novel dataset of manually annotated soccer penalty kicks to predict shot direction based on pre-kick player movements. We propose a deep learning classifier to benchmark this dataset that integrates HAR-based feature embeddings with contextual metadata. We evaluate twenty-two backbone models across seven architecture families (MViTv2, MViTv1, SlowFast, Slow, X3D, I3D, C2D), achieving up to 63.9% accuracy in predicting shot direction (left or right), outperforming the real goalkeepers' decisions. These results demonstrate the dataset's value for anticipatory action recognition and validate our model's potential as a generalizable approach for sports-based predictive tasks." }, "2507.19469.pdf": { "id": "2507.19469v1", "title": "Efficient Lines Detection for Robot Soccer", "year": "2025", "authors": [ "Jo\u00e3o G. Melo", "Jo\u00e3o P. Mafaldo", "Edna Barros" ], "categories": "cs.CV, cs.RO", "pdf_url": "https://arxiv.org/pdf/2507.19469v1", "abstract": "Self-localization is essential in robot soccer, where accurate detection of visual field features, such as lines and boundaries, is critical for reliable pose estimation. This paper presents a lightweight and efficient method for detecting soccer field lines using the ELSED algorithm, extended with a classification step that analyzes RGB color transitions to identify lines belonging to the field. We introduce a pipeline based on Particle Swarm Optimization (PSO) for threshold calibration to optimize detection performance, requiring only a small number of annotated samples. Our approach achieves accuracy comparable to a state-of-the-art deep learning model while offering higher processing speed, making it well-suited for real-time applications on low-power robotic platforms." }, "2508.04008.pdf": { "id": "2508.04008v1", "title": "Leveraging Minute-by-Minute Soccer Match Event Data to Adjust Team's Offensive Production for Game Context", "year": "2025", "authors": [ "Andrey Skripnikov", "Ahmet Cemek", "David Gillman" ], "categories": "stat.AP", "pdf_url": "https://arxiv.org/pdf/2508.04008v1", "abstract": "In soccer, game context can result in skewing offensive statistics in ways that might misrepresent how well a team has played. For instance, in England's 1-2 loss to France in the 2022 FIFA World Cup quarterfinal, England attempted considerably more shots (16 to France's 8) and more corners (5 to 2), potentially suggesting they played better despite the loss. However, these statistics were largely accumulated when France was ahead and more willing to concede offensive initiative to England. To explore how game context influences offensive performance, we analyze minute-by-minute event-sequenced match data from 15 seasons across five major European leagues. Using count-response Generalized Additive Modeling, we consider features such as score and red card differential, home/away status, pre-match win probabilities, and game minute. Moreover, we leverage interaction terms to test several intuitive hypotheses about how these features might cooperate in explaining offensive production. The selected model is then applied to project offensive statistics onto a standardized \"common denominator\" scenario: a tied home game with even men on both sides. The adjusted numbers - in contrast to regular game totals that disregard game context - offer a more contextualized comparison, reducing the likelihood of misrepresenting the relative quality of play." }, "2508.05891.pdf": { "id": "2508.05891v1", "title": "Bayesian weighted discrete-time dynamic models for association football prediction", "year": "2025", "authors": [ "Roberto Macr\u00ec-Demartino", "Leonardo Egidi", "Nicola Torelli" ], "categories": "stat.ME, stat.AP", "pdf_url": "https://arxiv.org/pdf/2508.05891v1", "abstract": "In recent years, great emphasis has been placed on the prediction of association football. Due to this, several studies have proposed different types of statistical models to predict the outcome of a football match. However, most existing approaches usually assume that the offensive and defensive abilities of teams remain static over time. We introduce a Bayesian dynamic approach for football goal based models that uses period-specific commensurate priors to flexibly weight the evolution of attacking and defensive abilities. Our approach assigns separate, time varying precisions for each ability and period, controlled via spike and slab hyperpriors. This adaptive shrinkage borrows information about teams' strength when past and current performance aligns and allows rapid adjustments when teams experience substantial changes (e.g., transfer windows or coaching changes). We integrate this framework into six standard goal based models evaluating predictive performance using data from the last five seasons of the German Bundesliga, English Premier League, and Spanish La Liga. Compared with the other discrete time dynamic models, our adaptive approach yields better predictive performance. The proposed methodology has also been implemented in the free and open source R package footBayes." }, "2508.07464.pdf": { "id": "2508.07464v1", "title": "Determining the acceleration field of a rigid body using three accelerometers and one gyroscope, with applications in mild traumatic brain injury", "year": "2025", "authors": [ "Yang Wan", "Benjamin E. Grossman-Ponemona", "Haneesh Kesari" ], "categories": "physics.app-ph", "pdf_url": "https://arxiv.org/pdf/2508.07464v1", "abstract": "Mild traumatic brain injury (mTBI) often results from violent head motion or impact. Most prevention strategies explicitly or implicitly rely on motion- or deformation-based injury criteria, both of which require accurate measurements of head motion. We present an algorithm for reconstructing the full acceleration field of a rigid body from measurements obtained by three tri-axial accelerometers and one tri-axial gyroscope. Unlike traditional gyroscope-based methods, which require numerically differentiating noisy angular velocity data, or gyroscope-free methods, which may impose restrictive sensor placement or involve nonlinear optimization, the proposed algorithm recovers angular acceleration and translational acceleration by solving a set of linear equations derived from rigid body kinematics. In the proposed method, the only constraint on sensor placement is that the accelerometers must be non-collinear. We validated the algorithm in controlled soccer heading experiments, demonstrating accurate prediction of accelerations at unsensed locations across trials. The proposed algorithm provides a robust, flexible, and efficient tool for reconstructing rigid body motion, with direct applications in contact sports, robotics, and biomechanical injury prediction." }, "2508.07543.pdf": { "id": "2508.07543v1", "title": "Commentary Generation for Soccer Highlights", "year": "2025", "authors": [ "Chidaksh Ravuru" ], "categories": "cs.CV, cs.LG", "pdf_url": "https://arxiv.org/pdf/2508.07543v1", "abstract": "Automated soccer commentary generation has evolved from template-based systems to advanced neural architectures, aiming to produce real-time descriptions of sports events. While frameworks like SoccerNet-Caption laid foundational work, their inability to achieve fine-grained alignment between video content and commentary remains a significant challenge. Recent efforts such as MatchTime, with its MatchVoice model, address this issue through coarse and fine-grained alignment techniques, achieving improved temporal synchronization. In this paper, we extend MatchVoice to commentary generation for soccer highlights using the GOAL dataset, which emphasizes short clips over entire games. We conduct extensive experiments to reproduce the original MatchTime results and evaluate our setup, highlighting the impact of different training configurations and hardware limitations. Furthermore, we explore the effect of varying window sizes on zero-shot performance. While MatchVoice exhibits promising generalization capabilities, our findings suggest the need for integrating techniques from broader video-language domains to further enhance performance. Our code is available at https://github.com/chidaksh/SoccerCommentary." }, "2508.09992.pdf": { "id": "2508.09992v1", "title": "OpenFPL: An open-source forecasting method rivaling state-of-the-art Fantasy Premier League services", "year": "2025", "authors": [ "Daniel Groos" ], "categories": "cs.LG, cs.AI", "pdf_url": "https://arxiv.org/pdf/2508.09992v1", "abstract": "Fantasy Premier League engages the football community in selecting the Premier League players who will perform best from gameweek to gameweek. Access to accurate performance forecasts gives participants an edge over competitors by guiding expectations about player outcomes and reducing uncertainty in squad selection. However, high-accuracy forecasts are currently limited to commercial services whose inner workings are undisclosed and that rely on proprietary data. This paper aims to democratize access to highly accurate forecasts of player performance by presenting OpenFPL, an open-source Fantasy Premier League forecasting method developed exclusively from public data. Comprising position-specific ensemble models optimized on Fantasy Premier League and Understat data from four previous seasons (2020-21 to 2023-24), OpenFPL achieves accuracy comparable to a leading commercial service when tested prospectively on data from the 2024-25 season. OpenFPL also surpasses the commercial benchmark for high-return players ($>$ 2 points), which are most influential for rank gains. These findings hold across one-, two-, and three-gameweek forecast horizons, supporting long-term planning of transfers and strategies while also informing final-day decisions." }, "2508.12930.pdf": { "id": "2508.12930v1", "title": "The path to a goal: Understanding soccer possessions via path signatures", "year": "2025", "authors": [ "David Hirnschall", "Robert Bajons" ], "categories": "stat.ML, cs.LG", "pdf_url": "https://arxiv.org/pdf/2508.12930v1", "abstract": "We present a novel framework for predicting next actions in soccer possessions by leveraging path signatures to encode their complex spatio-temporal structure. Unlike existing approaches, we do not rely on fixed historical windows and handcrafted features, but rather encode the entire recent possession, thereby avoiding the inclusion of potentially irrelevant or misleading historical information. Path signatures naturally capture the order and interaction of events, providing a mathematically grounded feature encoding for variable-length time series of irregular sampling frequencies without the necessity for manual feature engineering. Our proposed approach outperforms a transformer-based benchmark across various loss metrics and considerably reduces computational cost. Building on these results, we introduce a new possession evaluation metric based on well-established frameworks in soccer analytics, incorporating both predicted action type probabilities and action location. Our metric shows greater reliability than existing metrics in domain-specific comparisons. Finally, we validate our approach through a detailed analysis of the 2017/18 Premier League season and discuss further applications and future extensions." }, "2508.19477.pdf": { "id": "2508.19477v1", "title": "Concurrent validity of computer-vision artificial intelligence player tracking software using broadcast footage", "year": "2025", "authors": [ "Zachary L. Crang", "Rich D. Johnston", "Katie L. Mills", "Johsan Billingham", "Sam Robertson", "Michael H. Cole" ], "categories": "cs.CV, cs.AI", "pdf_url": "https://arxiv.org/pdf/2508.19477v1", "abstract": "This study aimed to: (1) understand whether commercially available computer-vision and artificial intelligence (AI) player tracking software can accurately measure player position, speed and distance using broadcast footage and (2) determine the impact of camera feed and resolution on accuracy. Data were obtained from one match at the 2022 Qatar Federation Internationale de Football Association (FIFA) World Cup. Tactical, programme and camera 1 feeds were used. Three commercial tracking providers that use computer-vision and AI participated. Providers analysed instantaneous position (x, y coordinates) and speed (m\\,s^{-1}) of each player. Their data were compared with a high-definition multi-camera tracking system (TRACAB Gen 5). Root mean square error (RMSE) and mean bias were calculated. Position RMSE ranged from 1.68 to 16.39 m, while speed RMSE ranged from 0.34 to 2.38 m\\,s^{-1}. Total match distance mean bias ranged from -1745 m (-21.8%) to 1945 m (24.3%) across providers. Computer-vision and AI player tracking software offer the ability to track players with fair precision when players are detected by the software. Providers should use a tactical feed when tracking position and speed, which will maximise player detection, improving accuracy. Both 720p and 1080p resolutions are suitable, assuming appropriate computer-vision and AI models are implemented." }, "2509.02722.pdf": { "id": "2509.02722v2", "title": "Planning with Reasoning using Vision Language World Model", "year": "2025", "authors": [ "Delong Chen", "Theo Moutakanni", "Willy Chung", "Yejin Bang", "Ziwei Ji", "Allen Bolourchi" ], "categories": "cs.AI", "pdf_url": "https://arxiv.org/pdf/2509.02722v2", "abstract": "Effective planning requires strong world models, but high-level world models that can understand and reason about actions with semantic and temporal abstraction remain largely underdeveloped. We introduce the Vision Language World Model (VLWM), a foundation model trained for language-based world modeling on natural videos. Given visual observations, the VLWM first infers the overall goal achievements then predicts a trajectory composed of interleaved actions and world state changes. Those targets are extracted by iterative LLM Self-Refine conditioned on compressed future observations represented by Tree of Captions. The VLWM learns both an action policy and a dynamics model, which respectively facilitates reactive system-1 plan decoding and reflective system-2 planning via cost minimization. The cost evaluates the semantic distance between the hypothetical future states given by VLWM roll-outs and the expected goal state, and is measured by a critic model that we trained in a self-supervised manner. The VLWM achieves state-of-the-art Visual Planning for Assistance (VPA) performance on both benchmark evaluations and our proposed PlannerArena human evaluations, where system-2 improves the Elo score by +27% upon system-1. The VLWM models also outperforms strong VLM baselines on RoboVQA and WorldPrediction benchmark." }, "2509.16095.pdf": { "id": "2509.16095v1", "title": "AdaSports-Traj: Role- and Domain-Aware Adaptation for Multi-Agent Trajectory Modeling in Sports", "year": "2025", "authors": [ "Yi Xu", "Yun Fu" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2509.16095v1", "abstract": "Trajectory prediction in multi-agent sports scenarios is inherently challenging due to the structural heterogeneity across agent roles (e.g., players vs. ball) and dynamic distribution gaps across different sports domains. Existing unified frameworks often fail to capture these structured distributional shifts, resulting in suboptimal generalization across roles and domains. We propose AdaSports-Traj, an adaptive trajectory modeling framework that explicitly addresses both intra-domain and inter-domain distribution discrepancies in sports. At its core, AdaSports-Traj incorporates a Role- and Domain-Aware Adapter to conditionally adjust latent representations based on agent identity and domain context. Additionally, we introduce a Hierarchical Contrastive Learning objective, which separately supervises role-sensitive and domain-aware representations to encourage disentangled latent structures without introducing optimization conflict. Experiments on three diverse sports datasets, Basketball-U, Football-U, and Soccer-U, demonstrate the effectiveness of our adaptive design, achieving strong performance in both unified and cross-domain trajectory prediction settings." }, "2509.20322.pdf": { "id": "2509.20322v2", "title": "VisualMimic: Visual Humanoid Loco-Manipulation via Motion Tracking and Generation", "year": "2025", "authors": [ "Shaofeng Yin", "Yanjie Ze", "Hong-Xing Yu", "C. Karen Liu", "Jiajun Wu" ], "categories": "cs.RO, cs.CV, cs.LG", "pdf_url": "https://arxiv.org/pdf/2509.20322v2", "abstract": "Humanoid loco-manipulation in unstructured environments demands tight integration of egocentric perception and whole-body control. However, existing approaches either depend on external motion capture systems or fail to generalize across diverse tasks. We introduce VisualMimic, a visual sim-to-real framework that unifies egocentric vision with hierarchical whole-body control for humanoid robots. VisualMimic combines a task-agnostic low-level keypoint tracker -- trained from human motion data via a teacher-student scheme -- with a task-specific high-level policy that generates keypoint commands from visual and proprioceptive input. To ensure stable training, we inject noise into the low-level policy and clip high-level actions using human motion statistics. VisualMimic enables zero-shot transfer of visuomotor policies trained in simulation to real humanoid robots, accomplishing a wide range of loco-manipulation tasks such as box lifting, pushing, football dribbling, and kicking. Beyond controlled laboratory settings, our policies also generalize robustly to outdoor environments. Videos are available at: https://visualmimic.github.io ." }, "2509.22522.pdf": { "id": "2509.22522v3", "title": "JointDiff: Bridging Continuous and Discrete in Multi-Agent Trajectory Generation", "year": "2025", "authors": [ "Guillem Capellera", "Luis Ferraz", "Antonio Rubio", "Alexandre Alahi", "Antonio Agudo" ], "categories": "cs.LG, cs.CV", "pdf_url": "https://arxiv.org/pdf/2509.22522v3", "abstract": "Generative models often treat continuous data and discrete events as separate processes, creating a gap in modeling complex systems where they interact synchronously. To bridge this gap, we introduce JointDiff, a novel diffusion framework designed to unify these two processes by simultaneously generating continuous spatio-temporal data and synchronous discrete events. We demonstrate its efficacy in the sports domain by simultaneously modeling multi-agent trajectories and key possession events. This joint modeling is validated with non-controllable generation and two novel controllable generation scenarios: weak-possessor-guidance, which offers flexible semantic control over game dynamics through a simple list of intended ball possessors, and text-guidance, which enables fine-grained, language-driven generation. To enable the conditioning with these guidance signals, we introduce CrossGuid, an effective conditioning operation for multi-agent domains. We also share a new unified sports benchmark enhanced with textual descriptions for soccer and football datasets. JointDiff achieves state-of-the-art performance, demonstrating that joint modeling is crucial for building realistic and controllable generative models for interactive systems. https://guillem-cf.github.io/JointDiff/" }, "2510.00480.pdf": { "id": "2510.00480v2", "title": "Expandable Decision-Making States for Multi-Agent Deep Reinforcement Learning in Soccer Tactical Analysis", "year": "2025", "authors": [ "Kenjiro Ide", "Taiga Someya", "Kohei Kawaguchi", "Keisuke Fujii" ], "categories": "cs.AI", "pdf_url": "https://arxiv.org/pdf/2510.00480v2", "abstract": "Invasion team sports such as soccer produce a high-dimensional, strongly coupled state space as many players continuously interact on a shared field, challenging quantitative tactical analysis. Traditional rule-based analyses are intuitive, while modern predictive machine learning models often perform pattern-matching without explicit agent representations. The problem we address is how to build player-level agent models from data, whose learned values and policies are both tactically interpretable and robust across heterogeneous data sources. Here, we propose Expandable Decision-Making States (EDMS), a semantically enriched state representation that augments raw positions and velocities with relational variables (e.g., scoring of space, pass, and score), combined with an action-masking scheme that gives on-ball and off-ball agents distinct decision sets. Compared to prior work, EDMS maps learned value functions and action policies to human-interpretable tactical concepts (e.g., marking pressure, passing lanes, ball accessibility) instead of raw coordinate features, and aligns agent choices with the rules of play. In the experiments, EDMS with action masking consistently reduced both action-prediction loss and temporal-difference (TD) error compared to the baseline. Qualitative case studies and Q-value visualizations further indicate that EDMS highlights high-risk, high-reward tactical patterns (e.g., fast counterattacks and defensive breakthroughs). We also integrated our approach into an open-source library and demonstrated compatibility with multiple commercial and open datasets, enabling cross-provider evaluation and reproducible experiments." }, "2510.01843.pdf": { "id": "2510.01843v1", "title": "Like Playing a Video Game: Spatial-Temporal Optimization of Foot Trajectories for Controlled Football Kicking in Bipedal Robots", "year": "2025", "authors": [ "Wanyue Li", "Ji Ma", "Minghao Lu", "Peng Lu" ], "categories": "cs.RO", "pdf_url": "https://arxiv.org/pdf/2510.01843v1", "abstract": "Humanoid robot soccer presents several challenges, particularly in maintaining system stability during aggressive kicking motions while achieving precise ball trajectory control. Current solutions, whether traditional position-based control methods or reinforcement learning (RL) approaches, exhibit significant limitations. Model predictive control (MPC) is a prevalent approach for ordinary quadruped and biped robots. While MPC has demonstrated advantages in legged robots, existing studies often oversimplify the leg swing progress, relying merely on simple trajectory interpolation methods. This severely constrains the foot's environmental interaction capability, hindering tasks such as ball kicking. This study innovatively adapts the spatial-temporal trajectory planning method, which has been successful in drone applications, to bipedal robotic systems. The proposed approach autonomously generates foot trajectories that satisfy constraints on target kicking position, velocity, and acceleration while simultaneously optimizing swing phase duration. Experimental results demonstrate that the optimized trajectories closely mimic human kicking behavior, featuring a backswing motion. Simulation and hardware experiments confirm the algorithm's efficiency, with trajectory planning times under 1 ms, and its reliability, achieving nearly 100 % task completion accuracy when the soccer goal is within the range of -90\u00b0 to 90\u00b0." }, "2510.04739.pdf": { "id": "2510.04739v1", "title": "ExposureEngine: Oriented Logo Detection and Sponsor Visibility Analytics in Sports Broadcasts", "year": "2025", "authors": [ "Mehdi Houshmand Sarkhoosh", "Fr\u00f8y \u00d8ye", "Henrik Nestor S\u00f8rlie", "Nam Hoang Vu", "Dag Johansen", "Cise Midoglu" ], "categories": "cs.CV, cs.MM", "pdf_url": "https://arxiv.org/pdf/2510.04739v1", "abstract": "Quantifying sponsor visibility in sports broadcasts is a critical marketing task traditionally hindered by manual, subjective, and unscalable analysis methods. While automated systems offer an alternative, their reliance on axis-aligned Horizontal Bounding Box (HBB) leads to inaccurate exposuremetrics when logos appear rotated or skewed due to dynamic camera angles and perspective distortions. This paper introduces ExposureEngine, an end-to-end system designed for accurate, rotation-aware sponsor visibility analytics in sports broadcasts, demonstrated in a soccer case study. Our approach predicts Oriented Bounding Box (OBB) to provide a geometrically precise fit to each logo regardless of the orientation on-screen. To train and evaluate our detector, we developed a new dataset comprising 1,103 frames from Swedish elite soccer, featuring 670 unique sponsor logos annotated with OBBs. Our model achieves a mean Average Precision (mAP@0.5) of 0.859, with a precision of 0.96 and recall of 0.87, demonstrating robust performance in localizing logos under diverse broadcast conditions. The system integrates these detections into an analytical pipeline that calculates precise visibility metrics, such as exposure duration and on-screen coverage. Furthermore, we incorporate a language-driven agentic layer, enabling users to generate reports, summaries, and media content through natural language queries. The complete system, including the dataset and the analytics dashboard, provides a comprehensive solution for auditable and interpretable sponsor measurement in sports media. An overview of the ExposureEngine is available online: https://youtu.be/tRw6OBISuW4 ." }, "2510.20287.pdf": { "id": "2510.20287v1", "title": "Breakdance Video classification in the age of Generative AI", "year": "2025", "authors": [ "Sauptik Dhar", "Naveen Ramakrishnan", "Michelle Munson" ], "categories": "cs.CV, cs.AI, cs.LG", "pdf_url": "https://arxiv.org/pdf/2510.20287v1", "abstract": "Large Vision Language models have seen huge application in several sports use-cases recently. Most of these works have been targeted towards a limited subset of popular sports like soccer, cricket, basketball etc; focusing on generative tasks like visual question answering, highlight generation. This work analyzes the applicability of the modern video foundation models (both encoder and decoder) for a very niche but hugely popular dance sports - breakdance. Our results show that Video Encoder models continue to outperform state-of-the-art Video Language Models for prediction tasks. We provide insights on how to choose the encoder model and provide a thorough analysis into the workings of a finetuned decoder model for breakdance video classification." }, "2510.23216.pdf": { "id": "2510.23216v4", "title": "Human-Like Goalkeeping in a Realistic Football Simulation: a Sample-Efficient Reinforcement Learning Approach", "year": "2025", "authors": [ "Alessandro Sestini", "Joakim Bergdahl", "Jean-Philippe Barrette-LaPierre", "Florian Fuchs", "Brady Chen", "Fabio Zinno" ], "categories": "cs.AI, cs.LG", "pdf_url": "https://arxiv.org/pdf/2510.23216v4", "abstract": "While several high profile video games have served as testbeds for Deep Reinforcement Learning (DRL), this technique has rarely been employed by the game industry for crafting authentic AI behaviors. Previous research focuses on training super-human agents with large models, which is impractical for game studios with limited resources aiming for human-like agents. This paper proposes a sample-efficient DRL method tailored for training and fine-tuning agents in industrial settings such as the video game industry. Our method improves sample efficiency of value-based DRL by leveraging pre-collected data and increasing network plasticity. We evaluate our method training a goalkeeper agent in EA SPORTS FC 25, one of the best-selling football simulations today. Our agent outperforms the game's built-in AI by 10% in ball saving rate. Ablation studies show that our method trains agents 50% faster compared to standard DRL methods. Finally, qualitative evaluation from domain experts indicates that our approach creates more human-like gameplay compared to hand-crafted agents. As a testament to the impact of the approach, the method has been adopted for use in the most recent release of the series." }, "2510.26389.pdf": { "id": "2510.26389v2", "title": "Adaptive Context Length Optimization with Low-Frequency Truncation for Multi-Agent Reinforcement Learning", "year": "2025", "authors": [ "Wenchang Duan", "Yaoliang Yu", "Jiwan He", "Yi Shi" ], "categories": "cs.LG, cs.MA", "pdf_url": "https://arxiv.org/pdf/2510.26389v2", "abstract": "Recently, deep multi-agent reinforcement learning (MARL) has demonstrated promising performance for solving challenging tasks, such as long-term dependencies and non-Markovian environments. Its success is partly attributed to conditioning policies on large fixed context length. However, such large fixed context lengths may lead to limited exploration efficiency and redundant information. In this paper, we propose a novel MARL framework to obtain adaptive and effective contextual information. Specifically, we design a central agent that dynamically optimizes context length via temporal gradient analysis, enhancing exploration to facilitate convergence to global optima in MARL. Furthermore, to enhance the adaptive optimization capability of the context length, we present an efficient input representation for the central agent, which effectively filters redundant information. By leveraging a Fourier-based low-frequency truncation method, we extract global temporal trends across decentralized agents, providing an effective and efficient representation of the MARL environment. Extensive experiments demonstrate that the proposed method achieves state-of-the-art (SOTA) performance on long-term dependency tasks, including PettingZoo, MiniGrid, Google Research Football (GRF), and StarCraft Multi-Agent Challenge v2 (SMACv2)." }, "2511.00121.pdf": { "id": "2511.00121v1", "title": "Analysis of Line Break prediction models for detecting defensive breakthrough in football", "year": "2025", "authors": [ "Shoma Yagi", "Jun Ichikawa", "Genki Ichinose" ], "categories": "cs.LG, physics.soc-ph, stat.AP", "pdf_url": "https://arxiv.org/pdf/2511.00121v1", "abstract": "In football, attacking teams attempt to break through the opponent's defensive line to create scoring opportunities. This action, known as a Line Break, is a critical indicator of offensive effectiveness and tactical performance, yet previous studies have mainly focused on shots or goal opportunities rather than on how teams break the defensive line. In this study, we develop a machine learning model to predict Line Breaks using event and tracking data from the 2023 J1 League season. The model incorporates 189 features, including player positions, velocities, and spatial configurations, and employs an XGBoost classifier to estimate the probability of Line Breaks. The proposed model achieved high predictive accuracy, with an AUC of 0.982 and a Brier score of 0.015. Furthermore, SHAP analysis revealed that factors such as offensive player speed, gaps in the defensive line, and offensive players' spatial distributions significantly contribute to the occurrence of Line Breaks. Finally, we found a moderate positive correlation between the predicted probability of being Line-Broken and the number of shots and crosses conceded at the team level. These results suggest that Line Breaks are closely linked to the creation of scoring opportunities and provide a quantitative framework for understanding tactical dynamics in football." }, "2511.00615.pdf": { "id": "2511.00615v1", "title": "Gaining Momentum: Uncovering Hidden Scoring Dynamics in Hockey through Deep Neural Sequencing and Causal Modeling", "year": "2025", "authors": [ "Daniel Griffiths", "Piper Moskow" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2511.00615v1", "abstract": "We present a unified, data-driven framework for quantifying and enhancing offensive momentum and scoring likelihood (expected goals, xG) in professional hockey. Leveraging a Sportlogiq dataset of 541,000 NHL event records, our end-to-end pipeline comprises five stages: (1) interpretable momentum weighting of micro-events via logistic regression; (2) nonlinear xG estimation using gradient-boosted decision trees; (3) temporal sequence modeling with Long Short-Term Memory (LSTM) networks; (4) spatial formation discovery through principal component analysis (PCA) followed by K-Means clustering on standardized player coordinates; and (5) use of an X-Learner causal inference estimator to quantify the average treatment effect (ATE) of adopting the identified \"optimal\" event sequences and formations. We observe an ATE of 0.12 (95% CI: 0.05-0.17, p < 1e-50), corresponding to a 15% relative gain in scoring potential. These results demonstrate that strategically structured sequences and compact formations causally elevate offensive performance. Our framework delivers real-time, actionable insights for coaches and analysts, advancing hockey analytics toward principled, causally grounded tactical optimization." }, "2511.03996.pdf": { "id": "2511.03996v2", "title": "Learning Vision-Driven Reactive Soccer Skills for Humanoid Robots", "year": "2025", "authors": [ "Yushi Wang", "Changsheng Luo", "Penghui Chen", "Jianran Liu", "Weijian Sun", "Tong Guo" ], "categories": "cs.RO", "pdf_url": "https://arxiv.org/pdf/2511.03996v2", "abstract": "Humanoid soccer poses a representative challenge for embodied intelligence, requiring robots to coordinate agile locomotion with unreliable visual perception in dynamic environments. However, existing systems typically rely on modular pipelines that separate perception from control or assume ideal sensing, making it difficult to achieve coherent and reactive behavior under real-world perceptual limitations. In this work, we present a unified reinforcement learning-based controller that enables humanoid robots to learn vision-driven reactive soccer skills by directly coupling visual perception with locomotion control. The robot is trained in simulation to acquire soccer behaviors, and adversarial motion priors guide policy learning toward natural motion patterns. To support robust performance under imperfect sensing, we introduce an encoder-decoder architecture together with a virtual perception system that models key characteristics of onboard vision, exposing the policy to perceptual noise and detection failures during training. This design encourages the policy to internalize perceptual uncertainty and continuously adapt its motion in a closed loop. The resulting controller produces coordinated soccer behaviors using only onboard vision, including ball searching, chasing, and multidirectional kicking. It reduces ball position estimation error by 46% and shortens time-to-kick by up to 64% compared with a rule-based baseline, achieving around 90% kicking success in frontfield positions. Experiments across diverse environments and dynamic scenarios, including real RoboCup competitions, further demonstrate the robust performance of the controller. These results highlight the practical effectiveness of integrating perceptual uncertainty directly into policy learning for achieving reliable vision-driven behaviors in humanoid robots operating under real-world conditions." }, "2511.06191.pdf": { "id": "2511.06191v1", "title": "Prediction-based evaluation of back-four defense with spatial control in soccer", "year": "2025", "authors": [ "Soujanya Dash", "Kenjiro Ide", "Rikuhei Umemoto", "Kai Amino", "Keisuke Fujii" ], "categories": "cs.CY, stat.AP", "pdf_url": "https://arxiv.org/pdf/2511.06191v1", "abstract": "Defensive organization is critical in soccer, particularly during negative transitions when teams are most vulnerable. The back-four defensive line plays a decisive role in preventing goal-scoring opportunities, yet its collective coordination remains difficult to quantify. This study introduces interpretable spatio-temporal indicators namely, space control, stretch index, pressure index, and defensive line height (absolute and relative) to evaluate the effectiveness of the back-four during defensive transitions. Using synchronized tracking and event data from the 2023-24 LaLiga season, 2,413 defensive sequences were analyzed following possession losses by FC Barcelona and Real Madrid CF. Two-way ANOVA revealed significant effects of team, outcome, and their interaction for key indicators, with relative line height showing the strongest association with defensive success. Predictive modeling using XGBoost achieved the highest discriminative performance (ROC AUC: 0.724 for Barcelona, 0.698 for Real Madrid), identifying space score and relative line height as dominant predictors. Comparative analysis revealed distinct team-specific defensive behaviors: Barcelona's success was characterized by higher spatial control and compact line coordination, whereas Real Madrid exhibited more adaptive but less consistent defensive structures. These findings demonstrate the tactical and predictive value of interpretable spatial indicators for quantifying collective defensive performance." }, "2511.07703.pdf": { "id": "2511.07703v2", "title": "Expected by Whom? A Skill-Adjusted Expected Goals Model for NHL Shooters and Goaltenders", "year": "2025", "authors": [ "J. T. P. Noel" ], "categories": "cs.OH", "pdf_url": "https://arxiv.org/pdf/2511.07703v2", "abstract": "This study outlines a light gradient boosted model aimed at predicting shot outcomes in the NHL. The model uses the NHL's spatiotemporal data to account for both the skill of shooters and goaltenders. This approach involves isolating and engineering features for different aspects of shooter and goaltender skill. These aspects include the overall skill, the locational skill, which is engineered using a shot binning technique previously outlined by Shuckers and Curro, and the situational skill, which is engineered using Gower distance. Three separate datasets were created based on the skill of the shooter and goaltender. For each, a baseline model was created in order to compare and contrast its performance with the skill-adjusted model. The results seen in this study show performance increases for the skill-adjusted model over the baseline model in log loss, brier scores, and area under the ROC curve. These performance increases have a high of 5\\% and outperform previous works, which have attempted to account only for player skill. This highlights the importance of accounting for both player and goaltender skill, while also accounting for different aspects of their skill. In future works, a skill-adjusted expected goals model could benefit models interested in predicting other aspects of the game, such as scoring leaders or individual game outcomes." }, "2511.09448.pdf": { "id": "2511.09448v1", "title": "MCAD: Multimodal Context-Aware Audio Description Generation For Soccer", "year": "2025", "authors": [ "Lipisha Chaudhary", "Trisha Mittal", "Subhadra Gopalakrishnan", "Ifeoma Nwogu", "Jaclyn Pytlarz" ], "categories": "cs.MM, cs.LG", "pdf_url": "https://arxiv.org/pdf/2511.09448v1", "abstract": "Audio Descriptions (AD) are essential for making visual content accessible to individuals with visual impairments. Recent works have shown a promising step towards automating AD, but they have been limited to describing high-quality movie content using human-annotated ground truth AD in the process. In this work, we present an end-to-end pipeline, MCAD, that extends AD generation beyond movies to the domain of sports, with a focus on soccer games, without relying on ground truth AD. To address the absence of domain-specific AD datasets, we fine-tune a Video Large Language Model on publicly available movie AD datasets so that it learns the narrative structure and conventions of AD. During inference, MCAD incorporates multimodal contextual cues such as player identities, soccer events and actions, and commentary from the game. These cues, combined with input prompts to the fine-tuned VideoLLM, allow the system to produce complete AD text for each video segment. We further introduce a new evaluation metric, ARGE-AD, designed to accurately assess the quality of generated AD. ARGE-AD evaluates the generated AD for the presence of five characteristics: (i) usage of people's names, (ii) mention of actions and events, (iii) appropriate length of AD, (iv) absence of pronouns, and (v) overlap from commentary or subtitles. We present an in-depth analysis of our approach on both movie and soccer datasets. We also validate the use of this metric to quantitatively comment on the quality of generated AD using our metric across domains. Additionally, we contribute audio descriptions for 100 soccer game clips annotated by two AD experts." }, "2511.12606.pdf": { "id": "2511.12606v3", "title": "Pixels or Positions? Benchmarking Modalities in Group Activity Recognition", "year": "2025", "authors": [ "Drishya Karki", "Merey Ramazanova", "Anthony Cioppa", "Silvio Giancola", "Bernard Ghanem" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2511.12606v3", "abstract": "Group Activity Recognition (GAR) is well studied on the video modality for surveillance and indoor team sports (e.g., volleyball, basketball). Yet, other modalities such as agent positions and trajectories over time, i.e. tracking, remain comparatively under-explored despite being compact, agent-centric signals that explicitly encode spatial interactions. Understanding whether pixel (video) or position (tracking) modalities leads to better group activity recognition is therefore important to drive further research on the topic. However, no standardized benchmark currently exists that aligns broadcast video and tracking data for the same group activities, leading to a lack of apples-to-apples comparison between these modalities for GAR. In this work, we introduce SoccerNet-GAR, a multimodal dataset built from the $64$ matches of the football World Cup 2022. Specifically, the broadcast videos and player tracking modalities for $87{,}939$ group activities are synchronized and annotated with $10$ categories. Furthermore, we define a unified evaluation protocol to benchmark two strong unimodal approaches: (i) competitive video-based classifiers and (ii) tracking-based classifiers leveraging graph neural networks. In particular, our novel role-aware graph architecture for tracking-based GAR directly encodes tactical structure through positional edges connecting players by their on-pitch roles. Our tracking model achieves $77.8\\%$ balanced accuracy compared to $60.9\\%$ for the best video baseline, while training with $7 \\times$ less GPU hours and $479 \\times$ fewer parameters ($180K$ vs. $86.3M$). This study provides new insights into the relative strengths of pixels and positions for group activity recognition in sports." }, "2511.14035.pdf": { "id": "2511.14035v2", "title": "Waiting for Dabo: A machine learning model for predicting Power 4 college football coaching hire success", "year": "2025", "authors": [ "Michael Schuckers", "Austin Hayes" ], "categories": "stat.AP", "pdf_url": "https://arxiv.org/pdf/2511.14035v2", "abstract": "Using data on 103 recent P4 college football hires, we built a statistical model for predicting a coach's success at their new school. For each hire, we collected data about their background and experiences, the previous success as a head coach or coordinator and their success since hiring. Over 50 variables on these factors were recorded though we used 29 of these in building our predictive model. Our measure of success is based upon Bill Connelly's SP+ team ratings relative to the performance on the same metric of the school in the 15 year prior to their selection as head coach. Using a cross-validated regularized linear regression, we obtain a predictive model for coaching success. Among the important factors for predicting a successful hire are having been a previous college head coach, leaving a job as an Offensive Coordinator, age and quality of the hiring school's team in the previous 15 years. While we do find these factors are important for the prediction of a successful coaching hire, the trends here are weak. With 66\\% accuracy, the model does identify coaching hires that will outperform team performance in the 15 years before the hire. However, no combination of these factors leads to high predictability of identifying a successful coaching hire. All of the data and code for this paper are available in a Github repository." }, "2511.15598.pdf": { "id": "2511.15598v2", "title": "Two families of reducible spherical conical metrics", "year": "2025", "authors": [ "Haoran Wu", "Xuwen Zhu" ], "categories": "math.DG", "pdf_url": "https://arxiv.org/pdf/2511.15598v2", "abstract": "We analyze a 1-parameter family of heart shape and a 3-parameter family obtained by gluing three footballs, both of which are examples of reducible spherical conical metrics. For these examples we verify the structure theorem given in [15] and show that such metrics naturally arise from Abelian differentials of the third kind. We then obtain the geometric decomposition using explicit metric and geodesic calculations. This offers new evidence for the interaction between the synthetic spherical geometry and the complex analytic structure of reducible conical metrics." }, "2511.16183.pdf": { "id": "2511.16183v1", "title": "FOOTPASS: A Multi-Modal Multi-Agent Tactical Context Dataset for Play-by-Play Action Spotting in Soccer Broadcast Videos", "year": "2025", "authors": [ "Jeremie Ochin", "Raphael Chekroun", "Bogdan Stanciulescu", "Sotiris Manitsaris" ], "categories": "cs.AI, cs.CV", "pdf_url": "https://arxiv.org/pdf/2511.16183v1", "abstract": "Soccer video understanding has motivated the creation of datasets for tasks such as temporal action localization, spatiotemporal action detection (STAD), or multiobject tracking (MOT). The annotation of structured sequences of events (who does what, when, and where) used for soccer analytics requires a holistic approach that integrates both STAD and MOT. However, current action recognition methods remain insufficient for constructing reliable play-by-play data and are typically used to assist rather than fully automate annotation. Parallel research has advanced tactical modeling, trajectory forecasting, and performance analysis, all grounded in game-state and play-by-play data. This motivates leveraging tactical knowledge as a prior to support computer-vision-based predictions, enabling more automated and reliable extraction of play-by-play data. We introduce Footovision Play-by-Play Action Spotting in Soccer Dataset (FOOTPASS), the first benchmark for play-by-play action spotting over entire soccer matches in a multi-modal, multi-agent tactical context. It enables the development of methods for player-centric action spotting that exploit both outputs from computer-vision tasks (e.g., tracking, identification) and prior knowledge of soccer, including its tactical regularities over long time horizons, to generate reliable play-by-play data streams. These streams form an essential input for data-driven sports analytics." }, "2511.18248.pdf": { "id": "2511.18248v2", "title": "Coherent Multi-Agent Trajectory Forecasting in Team Sports with CausalTraj", "year": "2025", "authors": [ "Wei Zhen Teoh" ], "categories": "cs.LG, cs.CV", "pdf_url": "https://arxiv.org/pdf/2511.18248v2", "abstract": "Jointly forecasting trajectories of multiple interacting agents is a core challenge in sports analytics and other domains involving complex group dynamics. Accurate prediction enables realistic simulation and strategic understanding of gameplay evolution. Most existing models are evaluated solely on per-agent accuracy metrics (minADE, minFDE), which assess each agent independently on its best-of-k prediction. However these metrics overlook whether the model learns which predicted trajectories can jointly form a plausible multi-agent future. Many state-of-the-art models are designed and optimized primarily based on these metrics. As a result, they may underperform on joint predictions and also fail to generate coherent, interpretable multi-agent scenarios in team sports. We propose CausalTraj, a temporally causal, likelihood-based model that is built to generate jointly probable multi-agent trajectory forecasts. To better assess collective modeling capability, we emphasize joint metrics (minJADE, minJFDE) that measure joint accuracy across agents within the best generated scenario sample. Evaluated on the NBA SportVU, Basketball-U, and Football-U datasets, CausalTraj achieves competitive per-agent accuracy and the best recorded results on joint metrics, while yielding qualitatively coherent and realistic gameplay evolutions." }, "2511.18730.pdf": { "id": "2511.18730v1", "title": "Large-Scale In-Game Outcome Forecasting for Match, Team and Players in Football using an Axial Transformer Neural Network", "year": "2025", "authors": [ "Michael Horton", "Patrick Lucey" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2511.18730v1", "abstract": "Football (soccer) is a sport that is characterised by complex game play, where players perform a variety of actions, such as passes, shots, tackles, fouls, in order to score goals, and ultimately win matches. Accurately forecasting the total number of each action that each player will complete during a match is desirable for a variety of applications, including tactical decision-making, sports betting, and for television broadcast commentary and analysis. Such predictions must consider the game state, the ability and skill of the players in both teams, the interactions between the players, and the temporal dynamics of the game as it develops. In this paper, we present a transformer-based neural network that jointly and recurrently predicts the expected totals for thirteen individual actions at multiple time-steps during the match, and where predictions are made for each individual player, each team and at the game-level. The neural network is based on an \\emph{axial transformer} that efficiently captures the temporal dynamics as the game progresses, and the interactions between the players at each time-step. We present a novel axial transformer design that we show is equivalent to a regular sequential transformer, and the design performs well experimentally. We show empirically that the model can make consistent and reliable predictions, and efficiently makes $\\sim$75,000 live predictions at low latency for each game." }, "2511.19885.pdf": { "id": "2511.19885v1", "title": "Complex Instruction Following with Diverse Style Policies in Football Games", "year": "2025", "authors": [ "Chenglu Sun", "Shuo Shen", "Haonan Hu", "Wei Zhou", "Chen Chen" ], "categories": "cs.MA, cs.LG", "pdf_url": "https://arxiv.org/pdf/2511.19885v1", "abstract": "Despite advancements in language-controlled reinforcement learning (LC-RL) for basic domains and straightforward commands (e.g., object manipulation and navigation), effectively extending LC-RL to comprehend and execute high-level or abstract instructions in complex, multi-agent environments, such as football games, remains a significant challenge. To address this gap, we introduce Language-Controlled Diverse Style Policies (LCDSP), a novel LC-RL paradigm specifically designed for complex scenarios. LCDSP comprises two key components: a Diverse Style Training (DST) method and a Style Interpreter (SI). The DST method efficiently trains a single policy capable of exhibiting a wide range of diverse behaviors by modulating agent actions through style parameters (SP). The SI is designed to accurately and rapidly translate high-level language instructions into these corresponding SP. Through extensive experiments in a complex 5v5 football environment, we demonstrate that LCDSP effectively comprehends abstract tactical instructions and accurately executes the desired diverse behavioral styles, showcasing its potential for complex, real-world applications." }, "2511.23072.pdf": { "id": "2511.23072v1", "title": "What If They Took the Shot? A Hierarchical Bayesian Framework for Counterfactual Expected Goals", "year": "2025", "authors": [ "Mikayil Mahmudlu", "Oktay Karaku\u015f", "Hasan Arkada\u015f" ], "categories": "eess.SP, cs.AI, stat.AP", "pdf_url": "https://arxiv.org/pdf/2511.23072v1", "abstract": "This study develops a hierarchical Bayesian framework that integrates expert domain knowledge to quantify player-specific effects in expected goals (xG) estimation, addressing a limitation of standard models that treat all players as identical finishers. Using 9,970 shots from StatsBomb's 2015-16 data and Football Manager 2017 ratings, we combine Bayesian logistic regression with informed priors to stabilise player-level estimates, especially for players with few shots. The hierarchical model reduces posterior uncertainty relative to weak priors and achieves strong external validity: hierarchical and baseline predictions correlate at R2 = 0.75, while an XGBoost benchmark validated against StatsBomb xG reaches R2 = 0.833. The model uncovers interpretable specialisation profiles, including one-on-one finishing (Aguero, Suarez, Belotti, Immobile, Martial), long-range shooting (Pogba), and first-touch execution (Insigne, Salah, Gameiro). It also identifies latent ability in underperforming players such as Immobile and Belotti. The framework supports counterfactual \"what-if\" analysis by reallocating shots between players under identical contexts. Case studies show that Sansone would generate +2.2 xG from Berardi's chances, driven largely by high-pressure situations, while Vardy-Giroud substitutions reveal strong asymmetry: replacing Vardy with Giroud results in a large decline (about -7 xG), whereas the reverse substitution has only a small effect (about -1 xG). This work provides an uncertainty-aware tool for player evaluation, recruitment, and tactical planning, and offers a general approach for domains where individual skill and contextual factors jointly shape performance." }, "2512.00203.pdf": { "id": "2512.00203v2", "title": "Beyond Expected Goals: A Probabilistic Framework for Shot Occurrences in Soccer", "year": "2025", "authors": [ "Jonathan Pipping-Gam\u00f3n", "Tianshu Feng", "R. Paul Sabin" ], "categories": "stat.AP, cs.LG, eess.IV", "pdf_url": "https://arxiv.org/pdf/2512.00203v2", "abstract": "Expected goals (xG) models estimate the probability that a shot results in a goal from its context (e.g., location, pressure), but they operate only on observed shots. We propose xG+, a possession-level framework that first estimates the probability that a shot occurs within the next second and its corresponding xG if it were to occur. We also introduce ways to aggregate this joint probability estimate over the course of a possession. By jointly modeling shot-taking behavior and shot quality, xG+ remedies the conditioning-on-shots limitation of standard xG. We show that this improves predictive accuracy at the team level and produces a more persistent player skill signal than standard xG models." }, "2512.04480.pdf": { "id": "2512.04480v8", "title": "Prescriptive Artificial Intelligence: A Formal Paradigm for Auditing Human Decisions Under Uncertainty", "year": "2025", "authors": [ "Pedro Passos" ], "categories": "cs.AI, cs.CE, eess.SY, math.OC", "pdf_url": "https://arxiv.org/pdf/2512.04480v8", "abstract": "We formalize Prescriptive Artificial Intelligence as a distinct paradigm for human-AI decision collaboration in high-stakes, stochastic environments involving single-agent individual decision-making. Unlike predictive systems optimized for outcome accuracy, prescriptive systems audit human decisions under uncertainty, providing normative guidance while preserving human agency and accountability. We introduce four domain-independent axioms characterizing prescriptive systems and prove fundamental separation results. Central is the Imitation Incompleteness theorem: supervised learning from historical decisions cannot correct systematic biases in the absence of external normative signals. Under standard regularity conditions, the induced predictor converges almost surely to the biased action rather than the normatively optimal one. Performance in decision imitation is therefore bounded by a structural bias term (epsilon_bias) rather than the statistical rate O(1/sqrt(n)), a result extended to Markovian logs and finite-sample concentration bounds. We demonstrate realizability through three independent instantiations spanning five decades: an interpretable fuzzy system for elite soccer auditing, revealing decision latency and risk states obscured by outcome and status quo biases; MYCIN, the historically validated rule-based clinical consultation system; and NEWS2, a nationally mandated clinical protocol validated on a prospective multi-center cohort. The framework establishes Prescriptive AI as a general, realizable class of decision-support systems for safety-critical domains where interpretability, contestability, and normative alignment are essential." }, "2512.08467.pdf": { "id": "2512.08467v1", "title": "Team-Aware Football Player Tracking with SAM: An Appearance-Based Approach to Occlusion Recovery", "year": "2025", "authors": [ "Chamath Ranasinghe", "Uthayasanker Thayasivam" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2512.08467v1", "abstract": "Football player tracking is challenged by frequent occlusions, similar appearances, and rapid motion in crowded scenes. This paper presents a lightweight SAM-based tracking method combining the Segment Anything Model (SAM) with CSRT trackers and jersey color-based appearance models. We propose a team-aware tracking system that uses SAM for precise initialization and HSV histogram-based re-identification to improve occlusion recovery. Our evaluation measures three dimensions: processing speed (FPS and memory), tracking accuracy (success rate and box stability), and robustness (occlusion recovery and identity consistency). Experiments on football video sequences show that the approach achieves 7.6-7.7 FPS with stable memory usage (~1880 MB), maintaining 100 percent tracking success in light occlusions and 90 percent in crowded penalty-box scenarios with 5 or more players. Appearance-based re-identification recovers 50 percent of heavy occlusions, demonstrating the value of domain-specific cues. Analysis reveals key trade-offs: the SAM + CSRT combination provides consistent performance across crowd densities but struggles with long-term occlusions where players leave the frame, achieving only 8.66 percent re-acquisition success. These results offer practical guidelines for deploying football tracking systems under resource constraints, showing that classical tracker-based methods work well with continuous visibility but require stronger re-identification mechanisms for extended absences." }, "2512.09431.pdf": { "id": "2512.09431v1", "title": "A Hierarchical, Model-Based System for High-Performance Humanoid Soccer", "year": "2025", "authors": [ "Quanyou Wang", "Mingzhang Zhu", "Ruochen Hou", "Kay Gillespie", "Alvin Zhu", "Shiqi Wang" ], "categories": "cs.RO", "pdf_url": "https://arxiv.org/pdf/2512.09431v1", "abstract": "The development of athletic humanoid robots has gained significant attention as advances in actuation, sensing, and control enable increasingly dynamic, real-world capabilities. RoboCup, an international competition of fully autonomous humanoid robots, provides a uniquely challenging benchmark for such systems, culminating in the long-term goal of competing against human soccer players by 2050. This paper presents the hardware and software innovations underlying our team's victory in the RoboCup 2024 Adult-Sized Humanoid Soccer Competition. On the hardware side, we introduce an adult-sized humanoid platform built with lightweight structural components, high-torque quasi-direct-drive actuators, and a specialized foot design that enables powerful in-gait kicks while preserving locomotion robustness. On the software side, we develop an integrated perception and localization framework that combines stereo vision, object detection, and landmark-based fusion to provide reliable estimates of the ball, goals, teammates, and opponents. A mid-level navigation stack then generates collision-aware, dynamically feasible trajectories, while a centralized behavior manager coordinates high-level decision making, role selection, and kick execution based on the evolving game state. The seamless integration of these subsystems results in fast, precise, and tactically effective gameplay, enabling robust performance under the dynamic and adversarial conditions of real matches. This paper presents the design principles, system architecture, and experimental results that contributed to ARTEMIS's success as the 2024 Adult-Sized Humanoid Soccer champion." }, "2512.11016.pdf": { "id": "2512.11016v2", "title": "SoccerMaster: A Vision Foundation Model for Soccer Understanding", "year": "2025", "authors": [ "Haolin Yang", "Jiayuan Rao", "Haoning Wu", "Weidi Xie" ], "categories": "cs.CV, cs.AI", "pdf_url": "https://arxiv.org/pdf/2512.11016v2", "abstract": "Soccer understanding has recently garnered growing research interest due to its domain-specific complexity and unique challenges. Unlike prior works that typically rely on isolated, task-specific expert models, this work aims to propose a unified model to handle diverse soccer visual understanding tasks, ranging from fine-grained perception (e.g., athlete detection and identification) to high-level semantic reasoning (e.g., event classification). Concretely, our contributions are threefold: (i) we present SoccerMaster, the first soccer-specific vision foundation model that unifies diverse tasks within a single framework via supervised multi-task pretraining; (ii) we develop an automated data curation pipeline, SoccerFactory, to generate scalable spatial annotations, and integrate multiple existing soccer video datasets as a comprehensive pretraining data resource for multi-task pretraining; and (iii) we conduct extensive evaluations demonstrating that SoccerMaster consistently outperforms task-specific expert models across diverse downstream tasks, highlighting its breadth and superiority. The data, code, and model will be publicly available." }, "2601.00748.pdf": { "id": "2601.00748v1", "title": "A Machine Learning Framework for Off Ball Defensive Role and Performance Evaluation in Football", "year": "2026", "authors": [ "Sean Groom", "Shuo Wang", "Francisco Belo", "Axl Rice", "Liam Anderson" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2601.00748v1", "abstract": "Evaluating off-ball defensive performance in football is challenging, as traditional metrics do not capture the nuanced coordinated movements that limit opponent action selection and success probabilities. Although widely used possession value models excel at appraising on-ball actions, their application to defense remains limited. Existing counterfactual methods, such as ghosting models, help extend these analyses but often rely on simulating \"average\" behavior that lacks tactical context. To address this, we introduce a covariate-dependent Hidden Markov Model (CDHMM) tailored to corner kicks, a highly structured aspect of football games. Our label-free model infers time-resolved man-marking and zonal assignments directly from player tracking data. We leverage these assignments to propose a novel framework for defensive credit attribution and a role-conditioned ghosting method for counterfactual analysis of off-ball defensive performance. We show how these contributions provide a interpretable evaluation of defensive contributions against context-aware baselines." }, "2601.10862.pdf": { "id": "2601.10862v1", "title": "Beyond Unidimensionality: General Factors and Residual Heterogeneity in Performance Evaluation", "year": "2026", "authors": [ "Krishna Sharma", "Pritam Basnet" ], "categories": "econ.GN, stat.AP", "pdf_url": "https://arxiv.org/pdf/2601.10862v1", "abstract": "How do evaluation systems compress multidimensional performance information into summary ratings? Using expert assessments of 9,669 professional soccer players on 28 attributes, we characterize the dimensional structure of evaluation outputs. The first principal component explains 40.6% of attribute variance, indicating a strong general factor, but formal noise discrimination procedures retain four components and bootstrap resampling confirms that this structure is highly stable. Internal consistency is high without evidence of redundancy. In out of sample prediction of expert overall ratings, a comprehensive model using the full attribute set substantially outperforms a single-factor summary (cross-validated R squared = 0.814). Overall, performance evaluations exhibit moderate information compression; they combine shared variance with stable residual dimensions that are economically meaningful for evaluation outcomes, with direct implications for the design of measurement systems." }, "2601.19479.pdf": { "id": "2601.19479v1", "title": "Time-to-Injury Forecasting in Elite Female Football: A DeepHit Survival Approach", "year": "2026", "authors": [ "Victoria Catterall", "Cise Midoglu", "Stephen Lynch" ], "categories": "cs.LG, cs.AI", "pdf_url": "https://arxiv.org/pdf/2601.19479v1", "abstract": "Injury occurrence in football poses significant challenges for athletes and teams, carrying personal, competitive, and financial consequences. While machine learning has been applied to injury prediction before, existing approaches often rely on static pre-season data and binary outcomes, limiting their real-world utility. This study investigates the feasibility of using a DeepHit neural network to forecast time-to-injury from longitudinal athlete monitoring data, while providing interpretable predictions. The analysis utilised the publicly available SoccerMon dataset, containing two seasons of training, match, and wellness records from elite female footballers. Data was pre-processed through cleaning, feature engineering, and the application of three imputation strategies. Baseline models (Random Forest, XGBoost, Logistic Regression) were optimised via grid search for benchmarking, while the DeepHit model, implemented with a multilayer perceptron backbone, was evaluated using chronological and leave-one-player-out (LOPO) validation. DeepHit achieved a concordance index of 0.762, outperforming baseline models and delivering individualised, time-varying risk estimates. Shapley Additive Explanations (SHAP) identified clinically relevant predictors consistent with established risk factors, enhancing interpretability. Overall, this study provides a novel proof of concept: survival modelling with DeepHit shows strong potential to advance injury forecasting in football, offering accurate, explainable, and actionable insights for injury prevention across competitive levels." }, "2601.21981.pdf": { "id": "2601.21981v1", "title": "VERSA: Verified Event Data Format for Reliable Soccer Analytics", "year": "2026", "authors": [ "Geonhee Jo", "Mingu Kang", "Kangmin Lee", "Minho Lee", "Pascal Bauer", "Sang-Ki Ko" ], "categories": "cs.AI, cs.DB", "pdf_url": "https://arxiv.org/pdf/2601.21981v1", "abstract": "Event stream data is a critical resource for fine-grained analysis across various domains, including financial transactions, system operations, and sports. In sports, it is actively used for fine-grained analyses such as quantifying player contributions and identifying tactical patterns. However, the reliability of these models is fundamentally limited by inherent data quality issues that cause logical inconsistencies (e.g., incorrect event ordering or missing events). To this end, this study proposes VERSA (Verified Event Data Format for Reliable Soccer Analytics), a systematic verification framework that ensures the integrity of event stream data within the soccer domain. VERSA is based on a state-transition model that defines valid event sequences, thereby enabling the automatic detection and correction of anomalous patterns within the event stream data. Notably, our examination of event data from the K League 1 (2024 season), provided by Bepro, detected that 18.81% of all recorded events exhibited logical inconsistencies. Addressing such integrity issues, our experiments demonstrate that VERSA significantly enhances cross-provider consistency, ensuring stable and unified data representation across heterogeneous sources. Furthermore, we demonstrate that data refined by VERSA significantly improves the robustness and performance of a downstream task called VAEP, which evaluates player contributions. These results highlight that the verification process is highly effective in increasing the reliability of data-driven analysis." }, "2602.05310.pdf": { "id": "2602.05310v1", "title": "Learning Soccer Skills for Humanoid Robots: A Progressive Perception-Action Framework", "year": "2026", "authors": [ "Jipeng Kong", "Xinzhe Liu", "Yuhang Lin", "Jinrui Han", "S\u00f6ren Schwertfeger", "Chenjia Bai" ], "categories": "cs.RO", "pdf_url": "https://arxiv.org/pdf/2602.05310v1", "abstract": "Soccer presents a significant challenge for humanoid robots, demanding tightly integrated perception-action capabilities for tasks like perception-guided kicking and whole-body balance control. Existing approaches suffer from inter-module instability in modular pipelines or conflicting training objectives in end-to-end frameworks. We propose Perception-Action integrated Decision-making (PAiD), a progressive architecture that decomposes soccer skill acquisition into three stages: motion-skill acquisition via human motion tracking, lightweight perception-action integration for positional generalization, and physics-aware sim-to-real transfer. This staged decomposition establishes stable foundational skills, avoids reward conflicts during perception integration, and minimizes sim-to-real gaps. Experiments on the Unitree G1 demonstrate high-fidelity human-like kicking with robust performance under diverse conditions-including static or rolling balls, various positions, and disturbances-while maintaining consistent execution across indoor and outdoor scenarios. Our divide-and-conquer strategy advances robust humanoid soccer capabilities and offers a scalable framework for complex embodied skill acquisition. The project page is available at https://soccer-humanoid.github.io/." }, "2602.12080.pdf": { "id": "2602.12080v2", "title": "PathCRF: Ball-Free Soccer Event Detection via Possession Path Inference from Player Trajectories", "year": "2026", "authors": [ "Hyunsung Kim", "Kunhee Lee", "Sangwoo Seo", "Sang-Ki Ko", "Jinsung Yoon", "Chanyoung Park" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2602.12080v2", "abstract": "Despite recent advances in AI, event data collection in soccer still relies heavily on labor-intensive manual annotation. Although prior work has explored automatic event detection using player and ball trajectories, ball tracking also remains difficult to scale due to high infrastructural and operational costs. As a result, comprehensive data collection in soccer is largely confined to top-tier competitions, limiting the broader adoption of data-driven analysis in this domain. To address this challenge, this paper proposes PathCRF, a framework for detecting on-ball soccer events using only player tracking data. We model player trajectories as a fully connected dynamic graph and formulate event detection as the problem of selecting exactly one edge corresponding to the current possession state at each time step. To ensure logical consistency of the resulting edge sequence, we employ a Conditional Random Field (CRF) that forbids impossible transitions between consecutive edges, where emission and transition scores are dynamically computed from edge embeddings produced by a socio-temporal backbone architecture. During inference, the most probable edge sequence is obtained via Viterbi decoding, and events such as ball controls or passes are detected whenever the selected edge changes between adjacent time steps. Experiments show that PathCRF produces accurate, logically consistent possession paths, enabling reliable downstream analyses while substantially reducing the need for manual event annotation. The source code is available at https://github.com/hyunsungkim-ds/pathcrf.git." }, "2602.15673.pdf": { "id": "2602.15673v1", "title": "Leicester's Tale: Another Perspective on the EPL 2015/16 Through Expected Goals (xG) Modelling", "year": "2026", "authors": [ "Sheikh Badar Ud Din Tahir", "Leonardo Egidi", "Nicola Torelli" ], "categories": "stat.ME", "pdf_url": "https://arxiv.org/pdf/2602.15673v1", "abstract": "Probabilistic modeling is an effective tool for evaluating team performance and predicting outcomes in sports. However, an important question that hasn't been fully explored is whether these models can reliably reflect actual performance while assigning meaningful probabilities to rare results that differ greatly from expectations. In this study, we create an inference-based probabilistic framework built on expected goals (xG). This framework converts shot-level event data into season-level simulations of points, rankings, and outcome probabilities. Using the English Premier League 2015/16 season as a data, we demonstrate that the framework captures the overall structure of the league table. It correctly identifies the top-four contenders and relegation candidates while explaining a significant portion of the variance in final points and ranks. In a full-season evaluation, the model assigns a low probability to extreme outcomes, particularly Leicester City's historic title win, which stands out as a statistical anomaly. We then look at the ex ante inferential and early-diagnostic role of xG by only using mid-season information. With first-half data, we simulate the rest of the season and show that teams with stronger mid-season xG profiles tend to earn more points in the second half, even after considering their current league position. In this mid-season assessment, Leicester City ranks among the top teams by xG and is given a small but noteworthy chance of winning the league. This suggests that their ultimate success was unlikely but not entirely detached from their actual performance. Our analysis indicates that expected goals models work best as probabilistic baselines for analysis and early-warning diagnostics, rather than as certain predictors of rare season outcomes." }, "2602.16830.pdf": { "id": "2602.16830v1", "title": "The Impact of Formations on Football Matches Using Double Machine Learning. Is it worth parking the bus?", "year": "2026", "authors": [ "Gen\u00eds Ruiz-Men\u00e1rguez", "Lloren\u00e7 Badiella" ], "categories": "stat.AP, cs.LG", "pdf_url": "https://arxiv.org/pdf/2602.16830v1", "abstract": "This study addresses a central tactical dilemma for football coaches: whether to employ a defensive strategy, colloquially known as \"parking the bus\", or a more offensive one. Using an advanced Double Machine Learning (DML) framework, this project provides a robust and interpretable tool to estimate the causal impact of different formations on key match outcomes such as goal difference, possession, corners, and disciplinary actions. Leveraging a dataset of over 22,000 matches from top European leagues, formations were categorized into six representative types based on tactical structure and expert consultation. A major methodological contribution lies in the adaptation of DML to handle categorical treatments, specifically formation combinations, through a novel matrix-based residualization process, allowing for a detailed estimation of formation-versus-formation effects that can inform a coach's tactical decision-making. Results show that while offensive formations like 4-3-3 and 4-2-3-1 offer modest statistical advantages in possession and corners, their impact on goals is limited. Furthermore, no evidence supports the idea that defensive formations, commonly associated with parking the bus, increase a team's winning potential. Additionally, red cards appear unaffected by formation choice, suggesting other behavioral factors dominate. Although this approach does not fully capture all aspects of playing style or team strength, it provides a valuable framework for coaches to analyze tactical efficiency and sets a precedent for future research in sports analytics." }, "2602.18519.pdf": { "id": "2602.18519v1", "title": "Wide Open Gazes: Quantifying Visual Exploratory Behavior in Soccer with Pose Enhanced Positional Data", "year": "2026", "authors": [ "Joris Bekkers" ], "categories": "cs.LG, cs.CV", "pdf_url": "https://arxiv.org/pdf/2602.18519v1", "abstract": "Traditional approaches to measuring visual exploratory behavior in soccer rely on counting visual exploratory actions (VEAs) based on rapid head movements exceeding 125\u00b0/s, but this method suffer from player position bias (i.e., a focus on central midfielders), annotation challenges, binary measurement constraints (i.e., a player is scanning, or not), lack the power to predict relevant short-term in-game future success, and are incompatible with fundamental soccer analytics models such as pitch control. This research introduces a novel formulaic continuous stochastic vision layer to quantify players' visual perception from pose-enhanced spatiotemporal tracking. Our probabilistic field-of-view and occlusion models incorporate head and shoulder rotation angles to create speed-dependent vision maps for individual players in a two-dimensional top-down plane. We combine these vision maps with pitch control and pitch value surfaces to analyze the awaiting phase (when a player is awaiting the ball to arrive after a pass for a teammate) and their subsequent on-ball phase. We demonstrate that aggregated visual metrics - such as the percentage of defended area observed while awaiting a pass - are predictive of controlled pitch value gained at the end of dribbling actions using 32 games of synchronized pose-enhanced tracking data and on-ball event data from the 2024 Copa America. This methodology works regardless of player position, eliminates manual annotation requirements, and provides continuous measurements that seamlessly integrate into existing soccer analytics frameworks. To further support the integration with existing soccer analytics frameworks we open-source the tools required to make these calculations." }, "2602.18981.pdf": { "id": "2602.18981v1", "title": "How Far Can We Go with Pixels Alone? A Pilot Study on Screen-Only Navigation in Commercial 3D ARPGs", "year": "2026", "authors": [ "Kaijie Xu", "Mustafa Bugti", "Clark Verbrugge" ], "categories": "cs.AI", "pdf_url": "https://arxiv.org/pdf/2602.18981v1", "abstract": "Modern 3D game levels rely heavily on visual guidance, yet the navigability of level layouts remains difficult to quantify. Prior work either simulates play in simplified environments or analyzes static screenshots for visual affordances, but neither setting faithfully captures how players explore complex, real-world game levels. In this paper, we build on an existing open-source visual affordance detector and instantiate a screen-only exploration and navigation agent that operates purely from visual affordances. Our agent consumes live game frames, identifies salient interest points, and drives a simple finite-state controller over a minimal action space to explore Dark Souls-style linear levels and attempt to reach expected goal regions. Pilot experiments show that the agent can traverse most required segments and exhibits meaningful visual navigation behavior, but also highlight that limitations of the underlying visual model prevent truly comprehensive and reliable auto-navigation. We argue that this system provides a concrete, shared baseline and evaluation protocol for visual navigation in complex games, and we call for more attention to this necessary task. Our results suggest that purely vision-based sense-making models, with discrete single-modality inputs and without explicit reasoning, can effectively support navigation and environment understanding in idealized settings, but are unlikely to be a general solution on their own." }, "2602.19582.pdf": { "id": "2602.19582v1", "title": "Advantage-based Temporal Attack in Reinforcement Learning", "year": "2026", "authors": [ "Shenghong He" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2602.19582v1", "abstract": "Extensive research demonstrates that Deep Reinforcement Learning (DRL) models are susceptible to adversarially constructed inputs (i.e., adversarial examples), which can mislead the agent to take suboptimal or unsafe actions. Recent methods improve attack effectiveness by leveraging future rewards to guide adversarial perturbation generation over sequential time steps (i.e., reward-based attacks). However, these methods are unable to capture dependencies between different time steps in the perturbation generation process, resulting in a weak temporal correlation between the current perturbation and previous perturbations.In this paper, we propose a novel method called Advantage-based Adversarial Transformer (AAT), which can generate adversarial examples with stronger temporal correlations (i.e., time-correlated adversarial examples) to improve the attack performance. AAT employs a multi-scale causal self-attention (MSCSA) mechanism to dynamically capture dependencies between historical information from different time periods and the current state, thus enhancing the correlation between the current perturbation and the previous perturbation. Moreover, AAT introduces a weighted advantage mechanism, which quantifies the effectiveness of a perturbation in a given state and guides the generation process toward high-performance adversarial examples by sampling high-advantage regions. Extensive experiments demonstrate that the performance of AAT matches or surpasses mainstream adversarial attack baselines on Atari, DeepMind Control Suite and Google football tasks." }, "2602.23233.pdf": { "id": "2602.23233v1", "title": "The Counterfactual Combine: A Causal Framework for Player Evaluation", "year": "2026", "authors": [ "Herbert P. Susmann", "Antonio D'Alessandro" ], "categories": "stat.AP", "pdf_url": "https://arxiv.org/pdf/2602.23233v1", "abstract": "Evaluating sports players based on their performance shares core challenges with evaluating healthcare providers based on patient outcomes. Drawing on recent advances in healthcare provider profiling, we cast sports player evaluation within a rigorous causal inference framework and define a flexible class of causal player evaluation estimands. Using stochastic interventions, we compare player success rates on repeated tasks (such as field goal attempts or plate appearance) to counterfactual success rates had those same attempts been randomly reassigned to players according to prespecified reference distributions. This setup encompasses direct and indirect standardization parameters familiar from healthcare provider profiling, and we additionally propose a \"performance above random replacement\" estimand designed for interpretability in sports settings. We develop doubly robust estimators for these evaluation metrics based on modern semiparametric statistical methods, with a focus on Targeted Minimum Loss-based Estimation, and incorporate machine learning methods to capture complex relationships driving player performance. We illustrate our framework in detailed case studies of field goal kickers in the National Football League and batters in Major League Baseball, highlighting how different causal estimands yield distinct interpretations and insights about player performance." }, "2603.00948.pdf": { "id": "2603.00948v1", "title": "HierKick: Hierarchical Reinforcement Learning for Vision-Guided Soccer Robot Control", "year": "2026", "authors": [ "Yizhi Chen", "Zheng Zhang", "Zhanxiang Cao", "Yihe Chen", "Shengcheng Fu", "Liyun Yan" ], "categories": "cs.RO", "pdf_url": "https://arxiv.org/pdf/2603.00948v1", "abstract": "Controlling soccer robots involves multi-time-scale decision-making, which requires balancing long-term tactical planning and short-term motion execution. Traditional end-to-end reinforcement learning (RL) methods face challenges in complex dynamic environments. This paper proposes HierKick, a vision-guided soccer robot control framework based on dual-frequency hierarchical RL. The framework adopts a hierarchical control architecture featuring a 5 Hz high-level policy that integrates YOLOv8 for real-time detection and selects tasks via a coach model, and a pre-trained 50 Hz low-level controller for precise joint control. Through this architecture, the framework achieves the four steps of approaching, aligning, dribbling, and kicking. Experimental results show that the success rates of this framework are 95.2\\% in IsaacGym, 89.8\\% in Mujoco, and 80\\% in the real world. HierKick provides an effective hierarchical paradigm for robot control in complex environments, extendable to multi-time-scale tasks, with its modular design and skill reuse offering a new path for intelligent robot control." }, "2603.06585.pdf": { "id": "2603.06585v1", "title": "Right Move, Right Time: Multi-Sport Space Evaluation Platform for Ultimate Frisbee, Basketball, and Soccer", "year": "2026", "authors": [ "Shunsuke Iwashita", "Titouan Jeannot", "Braden Eberhard", "Jacob Miller", "Rikako Kono", "Calvin Yeung" ], "categories": "cs.HC, cs.AI", "pdf_url": "https://arxiv.org/pdf/2603.06585v1", "abstract": "We present an open, sport-agnostic platform that turns tracking into comparable spatial measures across professional Ultimate, basketball, and soccer. Coaches in all three sports ask the same question: where is the usable space, and when should an off-ball run start? Our workflow standardizes inputs, provides timing-aware spatial evaluations, and makes it possible to reuse the same analysis across sports. We illustrate the approach with Ultimate as a focused testbed and then examine transfer between basketball and soccer. Together, these results show a practical path toward consistent, comparable evaluation across various invasion sports." }, "2603.11016.pdf": { "id": "2603.11016v3", "title": "A model-based restricted shapley value to measure the players' contribution to shot actions in football", "year": "2026", "authors": [ "Mattia Cefis", "Rodolfo Metulini", "Maurizio Carpita" ], "categories": "stat.AP", "pdf_url": "https://arxiv.org/pdf/2603.11016v3", "abstract": "This paper proposes a novel framework to assess individual player contributions in football, explicitly accounting for the cooperative nature of shot-ending offensive actions. By incorporating team interaction into player evaluation, it also supports economically sustainable decision-making, with practical implications for performance analysis and player scouting. Extending the expected Goals (xG) paradigm, we propose the expected Goal Action (xGA), an action-based metric designed to assess the quality of actions through passing networks. Furthermore, we adapt cooperative game theory and introduce the Player's Restricted Shapley Value (PRSV) statistic, a contribution metric based on restricted coalition structures derived from observed passing interactions, where xGA is adopted to compute the cohesion function. Unlike traditional Shapley approaches, the PRSV one restricts coalitions to tactically admissible player subsets, offering action-specific, interpretable measures of marginal contribution in a cooperative context. We apply the framework to 8,421 shot-actions from the Italian League Serie A season 2022/23, and the case studies of AC Milan and SSC Napoli reveal heterogeneity in contributions within teams. Combining the PRSV statistic with an individual performance metric highlights the discrepancies between a player's cooperative engagement and goal-conversion ability." }, "2603.15212.pdf": { "id": "2603.15212v2", "title": "Modeling Matches as Language: A Generative Transformer Approach for Counterfactual Player Valuation in Football", "year": "2026", "authors": [ "Miru Hong", "Minho Lee", "Geonhee Jo", "Hyeokje Cho", "Hyunsung Kim", "Pascal Bauer" ], "categories": "cs.AI, cs.LG", "pdf_url": "https://arxiv.org/pdf/2603.15212v2", "abstract": "Evaluating football player transfers is challenging because player actions depend strongly on tactical systems, teammates, and match context. Despite this complexity, recruitment decisions often rely on static statistics and subjective expert judgment, which do not fully account for these contextual factors. This limitation stems largely from the absence of counterfactual simulation mechanisms capable of predicting outcomes in hypothetical scenarios. To address these challenges, we propose ScoutGPT, a generative model that treats football match events as sequential tokens within a language modeling framework. Utilizing a NanoGPT-based Transformer architecture trained on next-token prediction, ScoutGPT learns the dynamics of match event sequences to simulate event sequences under hypothetical lineups, demonstrating superior predictive performance compared to existing baseline models. Leveraging this capability, the model employs Monte Carlo sampling to enable counterfactual simulation, allowing for the assessment of unobserved scenarios. Experiments on K League data show that simulated player transfers lead to measurable changes in offensive progression and goal probabilities, indicating that ScoutGPT captures player-specific impact beyond traditional static metrics." }, "2603.17687.pdf": { "id": "2603.17687v1", "title": "Objective Mispricing Detection for Shortlisting Undervalued Football Players via Market Dynamics and News Signals", "year": "2026", "authors": [ "Chinenye Omejieke", "Shuyao Chen", "Xia Cui" ], "categories": "cs.LG, cs.AI", "pdf_url": "https://arxiv.org/pdf/2603.17687v1", "abstract": "We present a practical, reproducible framework for identifying undervalued football players grounded in objective mispricing. Instead of relying on subjective expert labels, we estimate an expected market value from structured data (historical market dynamics, biographical and contract features, transfer history) and compare it to the observed valuation to define mispricing. We then assess whether news-derived Natural Language Processing (NLP) features (i.e., sentiment statistics and semantic embeddings from football articles) complement market signals for shortlisting undervalued players. Using a chronological (leakage-aware) evaluation, gradient-boosted regression explains a large share of the variance in log-transformed market value. For undervaluation shortlisting, ROC-AUC-based ablations show that market dynamics are the primary signal, while NLP features provide consistent, secondary gains that improve robustness and interpretability. SHAP analyses suggest the dominance of market trends and age, with news-derived volatility cues amplifying signals in high-uncertainty regimes. The proposed pipeline is designed for decision support in scouting workflows, emphasizing ranking/shortlisting over hard classification thresholds, and includes a concise reproducibility and ethics statement." }, "2603.27766.pdf": { "id": "2603.27766v1", "title": "AutoStan: Autonomous Bayesian Model Improvement via Predictive Feedback", "year": "2026", "authors": [ "Oliver D\u00fcrr" ], "categories": "cs.LG, stat.ML", "pdf_url": "https://arxiv.org/pdf/2603.27766v1", "abstract": "We present AutoStan, a framework in which a command-line interface (CLI) coding agent autonomously builds and iteratively improves Bayesian models written in Stan. The agent operates in a loop, writing a Stan model file, executing MCMC sampling, then deciding whether to keep or revert each change based on two complementary feedback signals: the negative log predictive density (NLPD) on held-out data and the sampler's own diagnostics (divergences, R-hat, effective sample size). We evaluate AutoStan on five datasets with diverse modeling structures. On a synthetic regression dataset with outliers, the agent progresses from naive linear regression to a model with Student-t robustness, nonlinear heteroscedastic structure, and an explicit contamination mixture, matching or outperforming TabPFN, a state-of-the-art black-box method, while remaining fully interpretable. Across four additional experiments, the same mechanism discovers hierarchical partial pooling, varying-slope models with correlated random effects, and a Poisson attack/defense model for soccer. No search algorithm, critic module, or domain-specific instructions are needed. This is, to our knowledge, the first demonstration that a CLI coding agent can autonomously write and iteratively improve Stan code for diverse Bayesian modeling problems." }, "2603.28916.pdf": { "id": "2603.28916v1", "title": "Structural Pass Analysis in Football: Learning Pass Archetypes and Tactical Impact from Spatio-Temporal Tracking Data", "year": "2026", "authors": [ "Oktay Karaku\u015f", "Hasan Arkada\u015f" ], "categories": "cs.LG, cs.CY", "pdf_url": "https://arxiv.org/pdf/2603.28916v1", "abstract": "The increasing availability of spatio-temporal tracking data has created new opportunities for analysing tactical behaviour in football. However, many existing approaches evaluate passes primarily through outcome-based metrics such as scoring probability or possession value, providing limited insight into how passes influence the defensive organisation of the opponent. This paper introduces a structural framework for analysing football passes based on their interaction with defensive structure. Using synchronised tracking/event data, we derive three complementary structural metrics, Line Bypass Score, Space Gain Metric, and Structural Disruption Index, that quantify how passes alter the spatial configuration of defenders. These metrics are combined into a composite measure termed Tactical Impact Value (TIV), which captures the structural influence of individual passes. Using tracking and event data from the 2022 FIFA World Cup, we analyse structural passing behaviour across multiple tactical levels. Unsupervised clustering of structural features reveals four interpretable pass archetypes: circulatory, destabilising, line-breaking, and space-expanding passes. Empirical results show that passes with higher TIV are significantly more likely to lead to territorial progression, particularly entries into the final third and penalty box. Spatial, team-level analyses further reveal distinctive structural passing styles across teams, while player-level analysis highlights the role of build-up defenders as key drivers of structural progression. In addition, analysing passer-receiver interactions identifies structurally impactful passing partnerships that amplify tactical progression within teams. Overall, the proposed framework demonstrates how structural representations derived from tracking data can reveal interpretable tactical patterns in football." }, "2604.00057.pdf": { "id": "2604.00057v1", "title": "Towards Automatic Soccer Commentary Generation with Knowledge-Enhanced Visual Reasoning", "year": "2026", "authors": [ "Zeyu Jin", "Xiaoyu Qin", "Songtao Zhou", "Kaifeng Yun", "Jia Jia" ], "categories": "cs.MM, cs.AI", "pdf_url": "https://arxiv.org/pdf/2604.00057v1", "abstract": "Soccer commentary plays a crucial role in enhancing the soccer game viewing experience for audiences. Previous studies in automatic soccer commentary generation typically adopt an end-to-end method to generate anonymous live text commentary. Such generated commentary is insufficient in the context of real-world live televised commentary, as it contains anonymous entities, context-dependent errors and lacks statistical insights of the game events. To bridge the gap, we propose GameSight, a two-stage model to address soccer commentary generation as a knowledge-enhanced visual reasoning task, enabling live-televised-like knowledgeable commentary with accurate reference to entities (players and teams). GameSight starts by performing visual reasoning to align anonymous entities with fine-grained visual and contextual analysis. Subsequently, the entity-aligned commentary is refined with knowledge by incorporating external historical statistics and iteratively updated internal game state information. Consequently, GameSight improves the player alignment accuracy by 18.5% on SN-Caption-test-align dataset compared to Gemini 2.5-pro. Combined with further knowledge enhancement, GameSight outperforms in segment-level accuracy and commentary quality, as well as game-level contextual relevance and structural composition. We believe that our work paves the way for a more informative and engaging human-centric experience with the AI sports application. Demo Page: https://gamesight2025.github.io/gamesight2025" }, "2604.05636.pdf": { "id": "2604.05636v1", "title": "Towards Athlete Fatigue Assessment from Association Football Videos", "year": "2026", "authors": [ "Xavier Bou", "Nathan Correger", "Alexandre Cloots", "C\u00e9dric Gavage", "Silvio Giancola", "C\u00e9dric Schwartz" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2604.05636v1", "abstract": "Fatigue monitoring is central in association football due to its links with injury risk and tactical performance. However, objective fatigue-related indicators are commonly derived from subjective self-reported metrics, biomarkers derived from laboratory tests, or, more recently, intrusive sensors such as heart monitors or GPS tracking data. This paper studies whether monocular broadcast videos can provide spatio-temporal signals of sufficient quality to support fatigue-oriented analysis. Building on state-of-the-art Game State Reconstruction methods, we extract player trajectories in pitch coordinates and propose a novel kinematics processing algorithm to obtain temporally consistent speed and acceleration estimates from reconstructed tracks. We then construct acceleration--speed (A-S) profiles from these signals and analyze their behavior as fatigue-related performance indicators. We evaluate the full pipeline on the public SoccerNet-GSR benchmark, considering both 30-second clips and a complete 45-minute half to examine short-term reliability and longer-term temporal consistency. Our results indicate that monocular GSR can recover kinematic patterns that are compatible with A-S profiling while also revealing sensitivity to trajectory noise, calibration errors, and temporal discontinuities inherent to broadcast footage. These findings support monocular broadcast video as a low-cost basis for fatigue analysis and delineate the methodological challenges for future research." }, "2604.06448.pdf": { "id": "2604.06448v2", "title": "From Load Tests to Live Streams: Graph Embedding-Based Anomaly Detection in Microservice Architectures", "year": "2026", "authors": [ "Srinidhi Madabhushi", "Pranesh Vyas", "Swathi Vaidyanathan", "Mayur Kurup", "Elliott Nash", "Yegor Silyutin" ], "categories": "cs.LG, cs.AI, cs.MM, eess.IV", "pdf_url": "https://arxiv.org/pdf/2604.06448v2", "abstract": "Prime Video regularly conducts load tests to simulate the viewer traffic spikes seen during live events such as Thursday Night Football as well as video-on-demand (VOD) events such as Rings of Power. While these stress tests validate system capacity, they can sometimes miss service behaviors unique to real event traffic. We present a graph-based anomaly detection system that identifies under-represented services using unsupervised node-level graph embeddings. Built on a GCN-GAE, our approach learns structural representations from directed, weighted service graphs at minute-level resolution and flags anomalies based on cosine similarity between load test and event embeddings. The system identifies incident-related services that are documented and demonstrates early detection capability. We also introduce a preliminary synthetic anomaly injection framework for controlled evaluation that show promising precision (96%) and low false positive rate (0.08%), though recall (58%) remains limited under conservative propagation assumptions. This framework demonstrates practical utility within Prime Video while also surfacing methodological lessons and directions, providing a foundation for broader application across microservice ecosystems." }, "2604.08722.pdf": { "id": "2604.08722v1", "title": "AI Driven Soccer Analysis Using Computer Vision", "year": "2026", "authors": [ "Adrian Manchado", "Tanner Cellio", "Jonathan Keane", "Yiyang Wang" ], "categories": "cs.CV, cs.AI", "pdf_url": "https://arxiv.org/pdf/2604.08722v1", "abstract": "Sport analysis is crucial for team performance since it provides actionable data that can inform coaching decisions, improve player performance, and enhance team strategies. To analyze more complex features from game footage, a computer vision model can be used to identify and track key entities from the field. We propose the use of an object detection and tracking system to predict player positioning throughout the game. To translate this to positioning in relation to the field dimensions, we use a point prediction model to identify key points on the field and combine these with known field dimensions to extract actual distances. For the player-identification model, object detection models like YOLO and Faster R-CNN are evaluated on the accuracy of our custom video footage using multiple different evaluation metrics. The goal is to identify the best model for object identification to obtain the most accurate results when paired with SAM2 (Segment Anything Model 2) for segmentation and tracking. For the key point detection model, we use a CNN model to find consistent locations in the soccer field. Through homography, the positions of points and objects in the camera perspective will be transformed to a real-ground perspective. The segmented player masks from SAM2 are transformed from camera perspective to real-world field coordinates through homography, regardless of camera angle or movement. The transformed real-world coordinates can be used to calculate valuable tactical insights including player speed, distance covered, positioning heatmaps, and more complex team statistics, providing coaches and players with actionable performance data previously unavailable from standard video analysis." }, "2604.13472.pdf": { "id": "2604.13472v2", "title": "Bridging MARL to SARL: An Order-Independent Multi-Agent Transformer via Latent Consensus", "year": "2026", "authors": [ "Zijian Zhao", "Jing Gao", "Sen Li" ], "categories": "cs.LG, cs.AI, cs.MA", "pdf_url": "https://arxiv.org/pdf/2604.13472v2", "abstract": "Cooperative multi-agent reinforcement learning (MARL) is widely used to address large joint observation and action spaces by decomposing a centralized control problem into multiple interacting agents. However, such decomposition often introduces additional challenges, including non-stationarity, unstable training, weak coordination, and limited theoretical guarantees. In this paper, we propose the Consensus Multi-Agent Transformer (CMAT), a centralized framework that bridges cooperative MARL to a hierarchical single-agent reinforcement learning (SARL) formulation. CMAT treats all agents as a unified entity and employs a Transformer encoder to process the large joint observation space. To handle the extensive joint action space, we introduce a hierarchical decision-making mechanism in which a Transformer decoder autoregressively generates a high-level consensus vector, simulating the process by which agents reach agreement on their strategies in latent space. Conditioned on this consensus, all agents generate their actions simultaneously, enabling order-independent joint decision making and avoiding the sensitivity to action-generation order in conventional Multi-Agent Transformers (MAT). This factorization allows the joint policy to be optimized using single-agent PPO while preserving expressive coordination through the latent consensus. To evaluate the proposed method, we conduct experiments on benchmark tasks from StarCraft II, Multi-Agent MuJoCo, and Google Research Football. The results show that CMAT achieves superior performance over recent centralized solutions, sequential MARL methods, and conventional MARL baselines. The code for this paper is available at:https://github.com/RS2002/CMAT ." }, "2604.16588.pdf": { "id": "2604.16588v1", "title": "MambaKick: Early Penalty Direction Prediction from HAR Embeddings", "year": "2026", "authors": [ "Henry O. Velesaca", "David Freire-Obregon", "Abel Reyes-Angulo", "Steven Araujo", "Angel Sappa" ], "categories": "cs.CV, cs.AI", "pdf_url": "https://arxiv.org/pdf/2604.16588v1", "abstract": "Penalty kicks in soccer are decided under extreme time constraints, where goalkeepers benefit from anticipating shot direction from the kickers motion before or around ball contact. In this paper, MambaKick is presented as a learning-based framework for penalty direction prediction that leverages pretrained human action recognition (HAR) embeddings extracted from contact-centered short video segments and combines them with a lightweight temporal predictor. Rather than relying on explicit kinematic reconstruction or handcrafted biomechanical features, the approach reuses transferable spatiotemporal representations and utilizes selective state-spare models (Mamba) for efficient sequence aggregation. Simple contextual metadata (e.g., field side and footedness) are also considered as complementary cues that may reduce ambiguity in real-world footage. Across a range of HAR backbones, MambaKick consistently improves or matches strong embedding baselines, achieving up to 53.1% accuracy for three classes and 64.5% for two classes under the proposed methodology. Overall, the results indicate that combining pretrained HAR representations with efficient state-space temporal modeling is a practical direction for low-latency intention prediction in real-world sports video. The code will be available at GitHub: https://github.com/hvelesaca/MambaKick/" }, "2604.17194.pdf": { "id": "2604.17194v1", "title": "Forecast Sports Outcomes under Efficient Market Hypothesis: Theoretical and Experimental Analysis of Odds-Only and Generalised Linear Models", "year": "2026", "authors": [ "Kaito Goto", "Naoya Takeishi", "Takehisa Yairi" ], "categories": "stat.ML, cs.LG", "pdf_url": "https://arxiv.org/pdf/2604.17194v1", "abstract": "Converting betting odds into accurate outcome probabilities is a fundamental challenge in order to use betting odds as a benchmark for sports forecasting and market efficiency analysis. In this study, we propose two methods to overcome the limitations of existing conversion methods. Firstly, we propose an odds-only method to convert betting odds to probabilities without using historical data for model fitting. While existing odds-only methods, such as Multiplicative, Shin, and Power exist, they do not adjust for biases or relationships we found in our betting odds dataset, which consists of 90014 football matches across five different bookmakers. To overcome these limitations, our proposed Odds-Only-Equal-Profitability-Confidence (OO-EPC) method aligns with the bookmakers' pricing objectives of having equal confidence in profitability for each outcome. We provide empirical evidence from our betting odds dataset that, for the majority of bookmakers, our proposed OO-EPC method outperforms the existing odds-only methods. Beyond controlled experiments, we applied the OO-EPC method under real-world uncertainty by using it for six iterations of an annual basketball outcome forecasting competition. Secondly, we propose a generalised linear model that utilises historical data for model fitting and then converts betting odds to probabilities. Existing generalised linear models attempt to capture relationships that the Efficient Market Hypothesis already captures. To overcome this shortcoming, our proposed Favourite-Longshot-Bias-Adjusted Generalised Linear Model (FL-GLM) fits just one parameter to capture the favourite-longshot bias, providing a more interpretable alternative. We provide empirical evidence from historical football matches where, for all bookmakers, our proposed FL-GLM outperforms the existing multinomial and logistic generalised linear models." }, "2604.18210.pdf": { "id": "2604.18210v1", "title": "TacticGen: Grounding Adaptable and Scalable Generation of Football Tactics", "year": "2026", "authors": [ "Sheng Xu", "Guiliang Liu", "Tarak Kharrat", "Yudong Luo", "Mohamed Aloulou", "Javier L\u00f3pez Pe\u00f1a" ], "categories": "cs.AI, cs.LG, cs.MA", "pdf_url": "https://arxiv.org/pdf/2604.18210v1", "abstract": "Success in association football relies on both individual skill and coordinated tactics. While recent advancements in spatio-temporal data and deep learning have enabled predictive analyses like trajectory forecasting, the development of tactical design remains limited. Bridging this gap is essential, as prediction reveals what is likely to occur, whereas tactic generation determines what should occur to achieve strategic objectives. In this work, we present TacticGen, a generative model for adaptable and scalable tactic generation. TacticGen formulates tactics as sequences of multi-agent movements and interactions conditioned on the game context. It employs a multi-agent diffusion transformer with agent-wise self-attention and context-aware cross-attention to capture cooperative and competitive dynamics among players and the ball. Trained with over 3.3 million events and 100 million tracking frames from top-tier leagues, TacticGen achieves state-of-the-art precision in predicting player trajectories. Building on it, TacticGen enables adaptable tactic generation tailored to diverse inference-time objectives through classifier guidance mechanism, specified via rules, natural language, or neural models. Its modeling performance is also inherently scalable. A case study with football experts confirms that TacticGen generates realistic, strategically valuable tactics, demonstrating its practical utility for tactical planning in professional football. The project page is available at: https://shengxu.net/TacticGen/." }, "2604.21087.pdf": { "id": "2604.21087v4", "title": "Model quality in football: Quantifying the quality of an Expected Threat model", "year": "2026", "authors": [ "Koen van Arem", "Jakob S\u00f6hl", "Mirjam Bruinsma", "Geurt Jongbloed" ], "categories": "stat.AP", "pdf_url": "https://arxiv.org/pdf/2604.21087v4", "abstract": "The recent growth in data availability in football has increased the risk of incorrect use of data-driven models, making guidelines on their validation and application necessary. The Expected Threat (xT) model is an accessible option for football organisations that start building in-house methods, yet little is known about how to assess its quality. The aim of this study is twofold: to examine how the model quality depends on the number of game states and the number of training points, and to translate these results into guidelines for constructing and applying the model. Using the Markov chain underlying the model, we perform theoretical analyses and simulations to study the estimation error. These show that the estimation error is approximately lognormal for a given number of training points and game states. Additionally, we combine the simulations with expert consultation to establish the estimation error beyond which player evaluations based on the Expected Threat model become unreliable for scouting applications. From this, we derive rules of thumb to ensure the quality of an Expected Threat model before application, and we illustrate through an example how a validated model can be applied in practice. Because the approach generalises to Expected Possession Value models, this paper illustrates a framework to systematically quantify model quality, despite the ground truth being unobservable in football analytics." }, "2604.23392.pdf": { "id": "2604.23392v1", "title": "SoccerRef-Agents: Multi-Agent System for Automated Soccer Refereeing", "year": "2026", "authors": [ "Zi Meng", "Wanli Song", "Yi Hu", "Jiayuan Rao", "Gang Chen" ], "categories": "cs.AI", "pdf_url": "https://arxiv.org/pdf/2604.23392v1", "abstract": "Refereeing is vital in sports, where fair, accurate, and explainable decisions are fundamental. While intelligent assistant technologies are being widely adopted in soccer refereeing, current AI-assisted approaches remain preliminary. Existing research mostly focuses on isolated video perception tasks and lacks the ability to understand and reason about foul scenarios. To fill this gap, we propose SoccerRef-Agents, a holistic and explainable multi-agent decision-making framework for soccer refereeing. The main contributions are: (i) constructing the multimodal benchmark SoccerRefBench with over 1,200 referee theory questions and 600 foul video clips; (ii) building a vector-based knowledge base RefKnowledgeDB using the latest \"Laws of the Game\" and a classic case database for precise, knowledge-driven reasoning; (iii) designing a novel multi-agent architecture that collaborates via cross-modal RAG to bridge the semantic gap between visual content and regulatory texts. This work explores the technical capability of integrating MLLMs with refereeing expertise, and evaluations show our system significantly outperforms general-purpose MLLMs in decision accuracy and explanation quality. All databases, benchmarks, and code will be made available." }, "2605.02623.pdf": { "id": "2605.02623v1", "title": "Retrieving Any Relevant Moments: Benchmark and Models for Generalized Moment Retrieval", "year": "2026", "authors": [ "Yiming Ding", "Siyu Cao", "Luyuan Jiao", "Yixuan Li", "Zitong Wang", "Zhiyong Liu" ], "categories": "cs.CV, cs.MM", "pdf_url": "https://arxiv.org/pdf/2605.02623v1", "abstract": "Video Moment Retrieval (VMR) aims to localize temporal segments in videos that correspond to a natural language query, but typically assumes only a single matching moment for each query. This assumption does not always hold in real-world scenarios, where queries may correspond to multiple or no moments. Thus, we formulate Generalized Moment Retrieval (GMR), a unified setting that requires retrieving the complete set of relevant moments or predicting an empty set. To enable systematic study of GMR, we introduce Soccer-GMR, a large-scale benchmark built on challenging soccer videos that reflect general GMR scenarios, with realistic negative and positive queries. The benchmark is constructed via a duration-flexible semi-automated pipeline with human verification, enabling scalable data generation while maintaining high annotation quality. We further design a unified evaluation protocol with complementary metrics tailored for null-set rejection, positive-query localization, and end-to-end GMR performance. Finally, we establish strong baselines across two modeling paradigms: a lightweight plug-and-play GMR adapter for discriminative VMR models, and a GMR-tailored GRPO reward for fine-tuning multimodal large language models (MLLMs). Extensive experiments show consistent gains across all metrics and expose key limitations of current methods, positioning GMR as a more realistic and challenging benchmark for video-language understanding." }, "2605.09598.pdf": { "id": "2605.09598v2", "title": "SoccerLens: Grounded Soccer Video Understanding Beyond Accuracy", "year": "2026", "authors": [ "Ismael Elsharkawi", "Ahmed Sait", "Silvio Giancola", "Bernard Ghanem", "Hossam Sharara", "Abdelrahman Eldesokey" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2605.09598v2", "abstract": "Vision-language models (VLMs) have recently shown strong potential in soccer video understanding. However, given the high complexity of soccer videos due to large viewpoint variations, rapid shot transitions, and cluttered scenes, it remains unclear on whether VLMs rely on meaningful visual evidence or exploit spurious correlations and shortcut learning. Existing evaluation protocols focus primarily on classification accuracy and do not assess visual grounding. To address this limitation, we introduce SoccerLens, a benchmark for grounded soccer video understanding. The benchmark contains annotated video segments spanning $13$ common soccer events, with structured visual cues organized into three levels of semantic relevance. We further extend the attribution method of Chefer [arXiv:2103.15679] to jointly model spatial and temporal attention, and introduce evaluation metrics that measure whether model attention aligns with annotated cues or drifts toward spurious regions. Our evaluation of state-of-the-art soccer VLMs shows that, despite strong classification accuracy, current models fail to exceed $50\\%$ grounding performance even under the loosest cue definitions and consistently underutilize temporal information. These results reveal a substantial gap between predictive performance and true visual grounding, highlighting the need for grounded evaluation in complex spatio-temporal domains such as soccer." }, "2605.10717.pdf": { "id": "2605.10717v1", "title": "Heteroscedastic Diffusion for Multi-Agent Trajectory Modeling", "year": "2026", "authors": [ "Guillem Capellera", "Antonio Rubio", "Luis Ferraz", "Antonio Agudo" ], "categories": "cs.LG, cs.CV", "pdf_url": "https://arxiv.org/pdf/2605.10717v1", "abstract": "Multi-agent trajectory modeling traditionally focuses on forecasting, often neglecting more general tasks like trajectory completion, which is essential for real-world applications such as correcting tracking data. Existing methods also generally predict agents' states without offering any state-wise measure of heteroscedastic uncertainty. Moreover, popular multi-modal sampling methods lack error probability estimates for each generated scene under the same prior observations, which makes it difficult to rank the predictions at inference time. We introduce U2Diffine, a unified diffusion model built to perform trajectory completion while simultaneously offering state-wise heteroscedastic uncertainty estimates. This is achieved by augmenting the standard denoising loss with the negative log-likelihood of the predicted noise, and then propagating the latent space uncertainty to the real state space using a first-order Taylor approximation. We also propose U2Diff, a faster baseline that avoids gradient computation during sampling. This approach significantly increases inference speed, making it as efficient as a standard generative-only diffusion model. For post-processing, we integrate a Rank Neural Network (RankNN) that enables error probability estimation for each generated mode, demonstrating strong correlation with ground truth errors. Our method outperforms state-of-the-art solutions in both trajectory completion and forecasting across four challenging sports datasets (NBA, Basketball-U, Football-U, Soccer-U), underscoring the effectiveness of our uncertainty and error probability estimation." }, "2605.10796.pdf": { "id": "2605.10796v1", "title": "Interpretable Machine Learning for Football Performance Analysis: Evidence of Limited Transferability from Elite Leagues to University Competition", "year": "2026", "authors": [ "Yu-Fang Tsai", "Yu-Jen Chen", "Kok-Hua Tan", "Sheng-Chieh Huang", "You-Ying Ji", "Yu-Lun Chen" ], "categories": "cs.AI", "pdf_url": "https://arxiv.org/pdf/2605.10796v1", "abstract": "Machine learning has become increasingly prevalent in football performance analysis, yet most studies prioritize predictive accuracy while implicitly assuming that learned performance determinants and their interpretations are transferable across competition levels. Whether interpretability remains reliable under domain shift-from elite to university football remains largely unexplored. This study investigates whether performance determinants learned from elite competitions are structurally transferable to university-level football and whether their interpretations remain robust under domain shift. Models were trained on large-scale event data from the top five European leagues and applied to university football data from National Tsing Hua University (NTHU) using an identical feature space. Random Forest and Multilayer Perceptron models were interpreted using SHapley Additive exPlanations (SHAP) and Counterfactual Impact Score (CIS). Across five experiments, elite football exhibited a stable and consistent hierarchy of performance determinants across leagues, models, and explanation methods. In contrast, NTHU university football showed substantial reordering of key indicators, reduced explanation stability, weaker structural agreement with elite domains, and increased sensitivity to explanation method. These findings suggest that interpretability robustness is domain-dependent. Rather than reflecting methodological limitations alone, instability in explanations under domain shift may serve as a diagnostic signal of structural ambiguity in the target domain." }, "2605.16066.pdf": { "id": "2605.16066v1", "title": "A market-calibrated accelerated failure time model for in-play football forecasting", "year": "2026", "authors": [ "Lawrence Clegg", "Zixing Song", "John Cartlidge" ], "categories": "stat.AP", "pdf_url": "https://arxiv.org/pdf/2605.16066v1", "abstract": "In-play football forecasting models have struggled to match the accuracy of betting exchange prices, which aggregate information from many market participants. We close this gap by combining two extensions to a Weibull accelerated failure time model: calibrating team strength parameters to Betfair Exchange prices at kick-off to capture pre-match market information, and including post-shot expected goals as a time-varying covariate to capture in-play information. The calibration approach, where we jointly fit team-strength parameters to 1X2 and over/under betting markets via squared-error minimisation, applies to any intensity-based goal arrival model and enables stronger in-play forecasting. Evaluated across 140 English Premier League matches at minute intervals, the calibrated model almost matches Betfair's classification accuracy (70.2% versus 70.6%) while retaining interpretable team-level parameters and covariate effects. A comparison with two alternative continuous-time scoring models, both calibrated to the same pre-match odds, confirms that market calibration is the dominant driver of predictive accuracy. A betting simulation against Betfair in-play odds yields a 4.5% return on investment (Sharpe ratio 5.94) over 17,458 bets, suggesting an inefficiency within in-play football markets." }, "2605.25696.pdf": { "id": "2605.25696v1", "title": "Evaluating passing decision-making in professional football: An enhanced MPNN approach to Receiver Selection", "year": "2026", "authors": [ "Gabriel Masella", "Giuseppe Alessio D'Inverno", "Max Goldsmith", "Gianluigi Rozza" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2605.25696v1", "abstract": "The process of decision-making in football is characterized by a complex interplay between spatial positioning, opponent pressure, and player intent. This work introduces a Graph Neural Network (GNN) framework designed to predict Receiver Selection, the optimal passing target, by modeling on-field interactions as dynamic graphs. Each player is represented as a node with positional and contextual features, while potential passing lines form weighted edges characterized by distance, angle, and pressure metrics. A Message-Passing Neural Network (MPNN) has been developed and trained using a combination of tracking data and event data from professional matches, synchronized through a robust pipeline based on an optimized version of the Needleman-Wunsch Algorithm. The model achieves competitive accuracy in identifying the actual chosen receiver and state-of-the-art accuracy within its top three suggestions. Our model further offers quantification of each option's likelihood, threat, and creativity, enabling performance analysts to evaluate over 1,000 passes in seconds." }, "2605.31529.pdf": { "id": "2605.31529v2", "title": "SVI-Bench: A Dynamic Microworld for Strategic Video Intelligence", "year": "2026", "authors": [ "Yulu Pan", "Han Yi", "Seongsu Ha", "Md Mohaiminul Islam", "Benjamin Zhang", "Lorenzo Torresani" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2605.31529v2", "abstract": "True video intelligence demands more than recognizing what is visible: it requires reasoning about why events unfold, predicting what would change under different conditions, and deciding what to do next. We refer to this progression, from perception through causal reasoning and simulation to strategic planning, as Strategic Video Intelligence (SVI). No existing benchmark evaluates this capability stack: in-the-wild videos lack verifiable ground truth for causal and strategic questions, while synthetic environments sacrifice the complexity of real multi-agent systems. To bridge this gap, we introduce SVI-Bench, a large-scale benchmark that leverages team sports as a dynamic microworld, combining the complexity of real-world multi-agent interaction (10-22 agents making coordinated decisions under adversarial pressure) with the verifiability of explicit rules and definitive outcomes. SVI-Bench comprises approximately 35K hours of broadcast video, 15M annotated actions, 15K hours of expert commentary, 23K game reports, and 103K structured statistical records across basketball, soccer, and hockey, all constructed via a data engine that transforms raw game data into a dense, cross-referenced corpus. We organize evaluation into 9 tasks spanning a progressive four-pillar hierarchy: Dynamic Scene Understanding, Causal Reasoning, Strategic Simulation, and Agentic Synthesis. Evaluating strong multimodal and agentic baselines, we find a capability cliff: models perform competently on perceptual tasks, achieving approximately 74% on fine-grained action QA, but degrade sharply at each successive cognitive level. Agentic tasks prove hardest: the strongest model achieves only 5% accuracy when required to autonomously gather and integrate evidence across a corpus of 1.8M clips." }, "2606.06353.pdf": { "id": "2606.06353v1", "title": "Maximising the Set-Piece Return: Optimising Football Corner Tactics with Graph Reinforcement Learning", "year": "2026", "authors": [ "Sean Groom", "Michael Groom", "Francisco Belo", "Axl Rice", "Liam Anderson", "Victor-Alexandru Darvariu" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2606.06353v1", "abstract": "Machine learning is increasingly employed for the evaluation of football tactics. However, existing approaches focus on characterising historical actions or analyst-specified counterfactual scenarios. In this work, we seek to go beyond the imitation of historically observed patterns towards discovering new generalisable player configurations and strategies. To tackle this, we focus on optimising corner kick routines, and formulate a decision-making problem in which a central policy makes adjustments to attacking player positions and velocities to maximise first contact shot probability. Unlike classic optimisation that solves for isolated setups, we contribute a reinforcement learning architecture operating on graph-structured data that yields a general policy for adjusting arbitrary starting player positions. Evaluated on over 3,000 Premier League corners, our approach strongly outperforms baseline optimisation techniques under matched inference budgets. Our results suggest that graph reinforcement learning can shift set-piece analysis from historical evaluation and imitation towards reward-driven tactical discovery." }, "2606.09289.pdf": { "id": "2606.09289v3", "title": "Intention Driven Identification of In-Possession Match Phases in Association Football through Temporal Graph Learning", "year": "2026", "authors": [ "Yuesen Li", "Daniel Link" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2606.09289v3", "abstract": "Understanding tactical organisation in association football requires identifying in-possession match phases that are shaped by evolving tactical intentions rather than by spatial patterns alone. This study proposes an intention-driven framework for identifying phases from tracking data. Seven German Bundesliga matches recorded at 25 Hz with TRACAB were analysed. A hierarchical model was defined with three tactical intentions (Invade Opponent Space, Keep Possession, Scoring) and six phases (Build Up, Progression, Counter Attack, Maintenance, Sustained Threat, Finishing). A Temporal Graph Attention Network (T-GAN) combined frame-level player-interaction graphs, contextual features, and Transformer-based temporal modelling. Performance was evaluated using frame-level F1 and temporal Intersection over Union (tIoU). T-GAN achieved macro-average frame-level F1 scores of 0.87 at the intention level, 0.76 for invasion-related phases, and 0.79 for scoring phases. After filtering, mean class-wise tIoU increased from 0.44 to 0.67 for intentions and from 0.40 to 0.57 for phases, showing that sequence-level evaluation captured fragmentation and boundary errors missed by frame-level metrics. Model comparisons indicated that Transformer-based sequence modelling drove coherent segmentation, while graph-based relational modelling was most beneficial for Counter Attack. Misalignment analysis revealed common sequence level misalignments, mainly Build Up/Progression ambiguity, Build Up/Maintenance confusion, Counter Attack/Progression inconsistency, and shortened pre-shot Finishing segments. Overall, the framework translates tracking data into tactically interpretable phase representations for automated annotation, tactical analysis, and playing-style profiling." }, "2606.09327.pdf": { "id": "2606.09327v1", "title": "A Universal Dense Football Event Representation Based on TabTransformer", "year": "2026", "authors": [ "Weiran Yang", "Daniel Memmert", "Maximilian Klemp-Weins" ], "categories": "cs.LG, cs.AI", "pdf_url": "https://arxiv.org/pdf/2606.09327v1", "abstract": "Football event data constitute a rich spatiotemporal source for quantitative analysis of player actions in team sports. These datasets contain heterogeneous features, combining continuous location coordinates with categorical variables such as action type, action outcome, and body part. Such data have been applied in sports analytics for match outcome forecasting, player evaluation, and tactical pattern recognition. However, existing approaches predominantly encode categorical features using one-hot or ordinal embedding representations, overlooking the intrinsic semantics of action descriptors. The Transformer is a deep neural network architecture based on self-attention that captures dependencies between input features at arbitrary positions. We propose and implement a Transformer-based model to learn latent dependencies among categorical event features and produce dense representations of football events. By encoding categorical features as learned embedding vectors, sport-specific action semantics are captured during pretraining, enabling the representations to support downstream tasks such as action value estimation and play style recognition. Empirical evaluation shows that the embedding representations yield superior probability calibration over task-specific baselines on the downstream prediction tasks, as measured by Brier score." }, "2606.09679.pdf": { "id": "2606.09679v1", "title": "SoccerNet 2026 Player-Centric Ball-Action Spotting:Retraining and Post-Processing Extensions to the FOOTPASS Baselines", "year": "2026", "authors": [ "Parthsarthi Rawat" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2606.09679v1", "abstract": "We describe our system for the SoccerNet 2026 Player-Centric Ball-Action Spotting Challenge, which requires predicting who performs which action and when, across eight classes in broadcast soccer. Building on the three FOOTPASS baselines [1] (TAAD, TAAD+GNN, and TAAD+DST), we contribute four extensions: (1) gradient check pointing to enable full-backbone fine-tuning on a single GPU; (2) fusion of GNN logits into the DST encoder, combining graph-based tactical context with per-player visual features; (3) square-root frequency class weighting to address the 213:1 pass-to-tackle imbalance in the training data; and (4) a post processing pipeline comprising per-class logit gating, temporal frame refinement, jersey re-assignment, and a two-model ensemble. Our system achieves 0.548 Macro F1 on the test set and 0.446 on the challenge set (server evaluation)." }, "2606.10568.pdf": { "id": "2606.10568v1", "title": "VeriSpace: Spatially Grounded Action Verification for Vision-Language-Action Models", "year": "2026", "authors": [ "Guiyu Zhao", "Longteng Guo", "Junyou Zhu", "Jun Fu", "Yanghong Mei", "Bin Cao" ], "categories": "cs.RO", "pdf_url": "https://arxiv.org/pdf/2606.10568v1", "abstract": "Vision-language-action (VLA) models have shown strong promise for robotic manipulation, but their reliability at test time remains limited by one-shot action prediction, where even small action errors can cause grasp failure, collision, or incorrect task progression. A natural alternative is to equip VLA systems with test-time verification, allowing multiple candidate actions to be proposed and evaluated before execution. However, reliable action verification is challenging because it requires not only distinguishing subtle geometric differences between candidate actions, but also assessing whether an action makes meaningful progress toward the task goal. We present VeriSpace, a 3D-aware action verifier for test-time action selection in VLA systems. VeriSpace evaluates candidate actions through two key components: Dual-Path 3D-Injected Scene Encoding, which constructs a scene representation that jointly preserves visual semantics and explicit 3D geometry, and Spatially-Grounded Action Reasoning, which evaluates each action by reasoning over task-relevant spatial relations, geometric validity, and expected goal progress. Together, these components enable more reliable discrimination between subtle yet outcome-critical action candidates while remaining fully compatible with existing VLA policies. Experiments on public benchmarks and real-world robotic manipulation tasks show that VeriSpace consistently improves decision reliability over both underlying VLA policies and prior verification-based methods, yielding substantial gains in both in-distribution and out-of-distribution settings." }, "2606.17056.pdf": { "id": "2606.17056v1", "title": "The Value Axis: Language Models Encode Whether They're on the Right Track", "year": "2026", "authors": [ "Nick Jiang", "Isaac Kauvar", "Jack Lindsey" ], "categories": "cs.CL", "pdf_url": "https://arxiv.org/pdf/2606.17056v1", "abstract": "We investigate whether language models internally track the value of their current trajectory, defined as the likelihood that their ongoing strategy will achieve their goals. Using synthetic, in-context reinforcement learning data, we construct a \"value\" axis for Qwen3-8B. We find that activations along this axis distinguish between high vs. low verbalized confidence, rollouts without and with backtracking, and correct vs. corrupted code. Steering towards high value causally suppresses self-correction and reduces explanatory verbosity, while steering towards low value induces backtracking and exploration. We demonstrate that direct preference optimization (DPO) can increase the internal value of rewarded behaviors (e.g. use a certain word), causing the model to act more confidently after exhibiting them. Finally, we apply the value axis to study in-the-wild settings. For example, we find that Qwen assigns low value to politically sensitive chat queries after post-training and that supervised fine-tuning increases internal confidence within the training domain. Our results suggest that language models linearly encode an estimate of expected goal success that modulates their confidence in pursuing a direction." }, "2606.18786.pdf": { "id": "2606.18786v2", "title": "R2D-RL: A RoboCup 2D Soccer Environment for Multi-Agent Reinforcement Learning", "year": "2026", "authors": [ "Haobin Qin", "Baofeng Zhang", "Hidehisa Akiyama", "Keisuke Fujii" ], "categories": "cs.AI", "pdf_url": "https://arxiv.org/pdf/2606.18786v2", "abstract": "Robot soccer is a challenging testbed for multi-agent reinforcement learning because it combines partial observability, cooperative and adversarial interaction, sparse rewards, and long-horizon tactical behavior. RoboCup 2D Soccer Simulation (RCSS2D) provides a mature robot-soccer platform, but its competition-oriented server-client architecture is difficult to use directly with modern Python-based MARL workflows. We introduce R2D-RL, a reinforcement learning environment that connects RCSS2D and HELIOS-based player clients to a Python MARL interface through shared-memory communication and cycle-level synchronization. R2D-RL supports full-field and scenario-based training with configurable opponents, Base discrete and Hybrid parameterized action spaces, action masks, expected possession value (EPV)-based reward shaping, and parallel execution. We provide front-goal scenarios and an 11-vs-11 full-field benchmark, together with baseline results." }, "2606.30946.pdf": { "id": "2606.30946v1", "title": "Head Kinematics and Brain Tissue Deformation from Soccer Heading: A Review of Implications for Brain Injury Risk", "year": "2026", "authors": [ "Christopher Lewis", "Anu Tripathi", "Alison Brooks", "Peter Ferrazzano", "Joseph Andrews", "Traci Snedden" ], "categories": "q-bio.OT", "pdf_url": "https://arxiv.org/pdf/2606.30946v1", "abstract": "Purpose: Repeated heading of soccer balls has raised concerns of potential long-term neurological effects. Consequently, numerous studies have estimated head kinematics and brain deformation due to soccer headers across different cohorts and play scenarios to identify higher risk conditions. However, heterogeneity in study design, data collection, and analysis has produced inconsistent findings, and injury risk is infrequently reported. Therefore, a meta-analysis of the existing literature was conducted to identify knowledge gaps and inform future studies assessing injury risk in soccer. Methods: We synthesized data from studies reporting head kinematics or brain deformation from soccer headers on human subjects. The data from these studies were analyzed to obtain the risk of mild traumatic brain injury (mTBI) based on applicable injury metrics and risk curves. Results: The meta-analysis revealed specific trends, indicating that match scenarios, corner and goal-kicks, top and oblique impacts, and older age cohorts were associated with higher head kinematics, while sex-based trends were inconclusive. The choice of sensor system affected the estimated head kinematics, with headband sensors consistently measuring higher kinematics than mouthpiece sensors. The data showed large variability stemming from heterogeneous study designs, limiting the applicability of the observed trends. These factors also influenced injury risk predictions, with estimated concussion risks generally below 20%. Conclusion: This review reveals trends in mTBI risk from soccer heading across different cohorts and play scenarios. It also underscores the need for standardized reporting of kinematics and brain deformation to enable mTBI risk estimation and meaningful cross-study comparisons." }, "2607.05845.pdf": { "id": "2607.05845v1", "title": "A behavioral principle underlying attacker-defender interactions in soccer", "year": "2026", "authors": [ "Issei Yamazaki", "Hirotaka Goto", "Kojiro Otoguro", "Hiraku Nishimori", "Masashi Shiraishi", "Takuma Narizuka" ], "categories": "physics.soc-ph", "pdf_url": "https://arxiv.org/pdf/2607.05845v1", "abstract": "Soccer is widely popular for its simple rules and complex yet coordinated play that unfolds on the pitch. Nevertheless, the fundamental mechanisms governing such play are not well understood: what shapes player interactions on the pitch? What short-term goals guide players' decisions about their movements over the next few seconds? We address these questions by focusing on one-on-one settings in open play, in which the attacker, in possession of the ball and typically dribbling, faces a defender aiming to stop or delay the attacker's actions over a short period. Here we develop a mathematical model of attacker-defender interactions and analyze 306 professional soccer games. Synthesizing the large-scale dataset with an analysis of the model reveals a simple behavioral principle that may underlie these interactions: the defender seeks to minimize their future relative speed to the attacker, whereas the attacker initiates their movements to preempt the defender's objective. This principle, relative-speed minimization, provides a consistent and unified account of the empirical data. Since our framework depends little on soccer-specific details, this principle may govern diverse pursuit-evasion scenarios as well as other invasion team sports." }, "2607.10990.pdf": { "id": "2607.10990v1", "title": "TreeSoc: Tree-Structured Dynamic Reasoning and Tool Synergy for Soccer Video Understanding", "year": "2026", "authors": [ "Thanh-Nhan Vo", "Thanh-Khoi Nguyen", "Trong-Thuan Nguyen", "Trung-Hoang Le", "Minh-Triet Tran" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2607.10990v1", "abstract": "Automated understanding of complex soccer scenarios from video remains a significant challenge for contemporary vision-language models (VLMs), which suffer from shallow cross-modal alignment and exhibit fundamental limitations in multi-step reasoning and coordinated tool integration. We present TreeSoc, a structured reasoning framework that reformulates soccer video question answering as a hierarchical search problem rather than a single-pass prediction. Specifically, TreeSoc employs a dynamic depth-first search (DFS) mechanism that decomposes complex queries into sequentially ordered sub-tasks, enabling iterative reasoning refinement through explicit intermediate states. This tree-structured decomposition naturally supports adaptive tool routing, wherein domain-specific modules are selectively activated and their outputs incorporated at each reasoning node to produce contextually grounded predictions. On SoccerBench, TreeSoc achieves state-of-the-art performance, with accuracies of 85.2%, 87.4%, and 82.2% on TextQA, ImageQA, and VideoQA, respectively. Additionally, TreeSoc further demonstrates strong cross-domain generalization, attaining 74.16% accuracy on NExT-QA. These results establish structured, tool-augmented tree reasoning as an effective paradigm for robust video understanding. Code is available at: https://github.com/thanhnhan29/TreeSoc." }, "2607.11548.pdf": { "id": "2607.11548v1", "title": "Training-Free Off-Screen Player Imputation for Broadcast-Based Spatial Football Analytics", "year": "2026", "authors": [ "Seongjin Choi" ], "categories": "cs.CV, cs.LG", "pdf_url": "https://arxiv.org/pdf/2607.11548v1", "abstract": "Spatial football metrics such as pitch control assume access to the positions of all 22 players, yet the most widely available source of positional data -- the broadcast main camera -- shows only 10-16 of them at any moment. We quantify the resulting distortion with an open, reproducible benchmark: a simulated broadcast viewport applied to open full-pitch tracking data (Metrica Sports; three matches, one held out from method development). Ignoring off-screen players -- the visible-only baseline implied whenever a video-based game-state-reconstruction (GSR) pipeline adds no imputation layer -- inflates hidden-zone pitch-control error to 25.1-26.9 percentage points and a mean absolute control-share error of 11.1-13.4 points across the three matches. We then evaluate a ladder of training-free, online imputation baselines that use only observations from the match being analysed. The best overall on these decision-relevant metrics, role-anchored centroid voting (each visible player votes for the full-team centroid by subtracting its running role offset, attenuating the viewport-induced subset bias), roughly halves hidden-zone error (to 12.2-13.8 points) and cuts control-share error to 28-48% of the ignore policy at every viewport width from 36 m to 60 m in all three matches. For occlusions <=9.6 s -- the regime of the closest learned prior work -- it reaches binwise median position errors of 3.3-8.9 m; but 50-57% of hidden-player observations lie beyond that regime. Integrated end-to-end into a broadcast-video GSR pipeline, imputation moves a downstream possession-quality score (Space-Creation Index) by 15.6 and 17.2 points on two real World Cup broadcast windows, flipping the verdict class in one." }, "2607.18084.pdf": { "id": "2607.18084v2", "title": "WorldCupArena: Fine-Grained Evaluation of Language Models and Deep-Research Agents on Football Forecasting", "year": "2026", "authors": [ "Zhaokai Wang", "Tianlin Gui", "Jiayuan Rao", "Shangzhe Di", "Yihong Tang", "Dingli Liang" ], "categories": "cs.AI, cs.CL, cs.LG", "pdf_url": "https://arxiv.org/pdf/2607.18084v2", "abstract": "Predicting a football match before kickoff requires more than knowing past results: a model must use changing information and make a clear prediction before the answer is available. We present WorldCupArena, a dynamic benchmark for language models and deep-research agents. The 2026 FIFA World Cup is its first evaluation, and the same process can be reused for future leagues and cups. Before each match, a model either receives a common evidence package or searches for information itself. It predicts the result and score, likely players and events, match statistics, and the outcome of the competition. After the match, these predictions are compared with the recorded result. We report result accuracy, exact-score accuracy, and a scoreline score that gives some credit when a predicted score is close but not exact, together with scores for the other prediction tasks. Across systems, similar result accuracy can mask larger differences in detailed predictions. Four systems predicted champion Spain, and two of them also recovered the exact final pairing. Compared with betting-market and human-fan baselines, the best system shows only small gains in result and exact-score accuracy, but a clearer gain in Scoreline. New schedules can be added as they begin, allowing the benchmark to evaluate future models without using outcomes that are already known. Code, predictions and evaluation scripts will be publicly released." }, "2607.19783.pdf": { "id": "2607.19783v2", "title": "Do In-Match Hydration Breaks Alter Match Momentum? A Within-Match Case-Crossover Analysis of the 2026 FIFA World Cup", "year": "2026", "authors": [ "Debangan Dey" ], "categories": "stat.AP, stat.ME", "pdf_url": "https://arxiv.org/pdf/2607.19783v2", "abstract": "The 2026 FIFA World Cup was the first to mandate hydration breaks, brief stoppages near the twenty-second minute of each half, in every match regardless of weather. A widespread claim holds that these breaks blunt the momentum of whichever side is dominant. We test the claim with a within-match case-crossover design, using non-break minutes of the same match as self-matched controls. This differences out every match-level characteristic and adjusts only for game time, scoreline, and pre-break momentum state. The apparent post-break fade is not caused by the break: the dominant side's momentum decays at essentially the same rate whether or not a break occurs, reverting through the stoppage as if play had continued. What the break removes instead is the continuation of that side's pressure during the stopped minutes, and this leaves no trace in net expected goals. The average effect is only weakly identified, resting on a trend extrapolation into a window with no untreated minutes, whereas break-by-covariate interactions are stable; that average is small and near zero. With only 99 matches the design lacks power to reject a small effect, so we read the results by effect size and confidence interval. The estimates are not uniform across match states. The clearest modifier is team strength: the break effect shifts by about +1.6 momentum points per 100 Elo points of rating advantage, favouring a stronger in-form side over a weaker side that is momentarily dominant, and it is the only statistically significant interaction. A weaker pattern appears by scoreline, where a comfortably-leading side of average strength shows an implied loss near -4 points on a pre-break level of about +20. We report these state-dependent signals as exploratory and in need of replication." }, "2607.26061.pdf": { "id": "2607.26061v1", "title": "Sim2Win: A Team-Agnostic, Event-Based Pre-Match Outcome Prediction and Tactical Profiling System for Football", "year": "2026", "authors": [ "Mouad Zemzoumi", "Amine Abouaomar" ], "categories": "cs.LG, cs.AI", "pdf_url": "https://arxiv.org/pdf/2607.26061v1", "abstract": "Pre-match tactical decision-making in professional football relies heavily on subjective expert analysis and identity-based scouting systems that cannot generalize to unseen teams. This paper presents Sim2Win, a team-agnostic, event-based pre-match tactical recommendation framework that reframes match outcome prediction as a tactical decision-support problem. Using StatsBomb open event data from eleven competitions spanning 178 teams and 1,411 team-match records, Sim2Win constructs five-match rolling tactical profiles, engineers four interpretable tactical feature ratios, clusters team behaviors into eight playstyles via K-Means, and trains thirteen classifiers to estimate win, draw, and loss probabilities from tactical matchup representations. The system operates without team names or identity features, enabling generalization to teams never seen during training. A rigorous Leave-One-Competition-Out (LOCO) evaluation demonstrates that Sim2Win achieves a mean ROC-AUC of 0.704 and mean accuracy of 55.4% on completely unseen teams, outperforming ELO, Pi-Rating, and GAP baselines on all 21 ROC-AUC comparisons and 19 of 21 accuracy comparisons. Among all evaluated models, CatBoost achieved the strongest in-distribution performance with 60.90% accuracy. These findings suggest that behavioral tactical representations provide transferable predictive signal under distribution shift and offer a viable alternative to identity-dependent football prediction systems." }, "2607.27574.pdf": { "id": "2607.27574v1", "title": "Policy Gradient Steering: Interventions from Behavioral Objectives", "year": "2026", "authors": [ "Yoann Poupart", "Aur\u00e9lie Beynier", "Nicolas Maudet" ], "categories": "cs.LG, cs.MA", "pdf_url": "https://arxiv.org/pdf/2607.27574v1", "abstract": "Activation steering has emerged in large language models as a lightweight alternative for dynamically changing a model's behavior at inference time. However, we show that existing steering methods fail to steer even a simple policy in a two-route gridworld environment. To address this limitation, we propose Policy Gradient Steering (PGS), which formulates steering as a reinforcement learning problem. PGS accumulates gradients of a temporary behavioral objective over a small set of rollouts or demonstrations to construct a removable task vector. We first demonstrate the calibration and reversibility of PGS in a two-route gridworld environment. Using chess puzzles, we then evaluate independently fitted PGS vectors both in isolation and in combination, finding that compatible tactical objectives accumulate constructively. Finally, in competitive football, we show that PGS can alter specific team behaviors and that its effects transfer across opponents. Together, these results show that policy gradients provide a natural interface for constructing temporary and composable behavioral adaptations across diverse decision-making domains." }, "2607.28531.pdf": { "id": "2607.28531v1", "title": "AIx4Soccer: A Unified Platform Architecture for Football Club Management and Structured Athlete Development", "year": "2026", "authors": [ "Frederico Falconi Costa", "Salvador Cesar Costa", "Fabricio F. Costa" ], "categories": "cs.CY", "pdf_url": "https://arxiv.org/pdf/2607.28531v1", "abstract": "Football clubs, academies, and federations operate a growing but fragmented portfolio of digital tools: separate systems for video analysis, GPS/performance tracking, medical records, scouting, and administration. This fragmentation is most acute outside the elite European clubs that can afford integration, producing a digital divide that disadvantages grassroots clubs in developing markets such as Brazil, paradoxically the world's largest exporter of professional players. This paper presents, at a conceptual level, the architecture of \"AIx4Soccer One Platform,\" a multi-tenant cloud SaaS operating system that unifies club-management workflows and embeds a structured athlete-development methodology, the PDI Framework (Plano de Desenvolvimento Individual / Individual Development Plan). We describe two companion components: \"Tak Tik,\" a certified two-sided marketplace connecting clubs with video analysts under a 75%/25% (analyst/platform) revenue split, and the PDI/TBIL methodology, which links development plans to video evidence and periodic review. As Materials and Methods, we state explicit requirements and give a formal, implementation-independent specification of the platform's proposed future substrate: an event-centric semantic data model in which every fact is a typed, immutable event in an append-only log that induces a growing knowledge graph. We situate the design against the literature on athlete-development frameworks, sports-analytics workflows, two-sided-market economics, and multi-tenant SaaS patterns, and discuss youth data-protection obligations (Brazil's LGPD and 2025 Digital ECA; the EU GDPR), algorithmic-fairness risks in talent evaluation, and why small, domain-specific models, rather than frontier LLMs, are the appropriate intelligence layer. This is a design and early-deployment paper, not an empirical evaluation, making no efficacy claims." }, "2608.02081.pdf": { "id": "2608.02081v1", "title": "Isotonic Bradley-Terry Model for Paired Comparison Data", "year": "2026", "authors": [ "Ryoya Yamasaki" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2608.02081v1", "abstract": "In this paper, we study prediction problems for paired comparison data, for example, predicting the win probability between two unmatched players and ranking all the players according to the order of their strengths by using win probability data between two matched players. Paired comparison data are typically analyzed using Bradley-Terry and Thurstone-Mosteller models. These models predict the win probability by transforming the difference between learned rate parameters, which represent players'\\;strengths, with a pre-specified inverse link function, and employ the order of learned rate parameters for player ranking. However, these models may suffer from model misspecification owing to the selection of a fixed inverse link function. Therefore, in this study, we propose to learn the rate parameters by a (sub-)gradient method and the inverse link function by an isotonic regression technique alternately. The proposed model guarantees monotonic improvement in training error, and is likely to yield an exact tie when the available data is insufficient to establish a strict ranking. We also verified that the proposed model could improve the win probability prediction and ranking performance through numerical experiments with synthetic data and real-world data of football Premier League, baseball MLB, and tennis ATP tour." }, "2608.03416.pdf": { "id": "2608.03416v1", "title": "AI World Cup 2026: Benchmarking Large Language Models for End-to-End Football Tournament Prediction", "year": "2026", "authors": [ "Jonaid Shianifar", "Iias Faiud" ], "categories": "cs.AI, cs.LG", "pdf_url": "https://arxiv.org/pdf/2608.03416v1", "abstract": "Large language models (LLMs) are now regularly asked to forecast real-world events, but comparisons are often difficult because models receive different information, use different tools, and are evaluated under different rules. This paper reports the completed \\emph{AI World Cup} benchmark, in which ten LLM-based assistants made a single pre-tournament forecast of the entire 2026 FIFA World Cup. Every submission used the same tournament snapshot, prompt, JSON schema, and scoring procedure. The forecasts covered group-stage scores, group rankings, the knockout bracket, final placings, confidence values, and short explanations. After all 104 matches had been played, GPT-5.5 Thinking finished first with 744 points, followed by GPT-5.5 with 717, Gemini with 699, and Qwen 3.7 with 687. GPT-5.5 Thinking was also the only model to select Spain, which defeated Argentina 1--0 in the final, as champion. The final ranking was driven mainly by knockout performance: total score was strongly correlated with knockout points ($r=0.986$), but showed little relationship with group-stage match points ($r=0.055$), group-standing points ($r=-0.103$), or their combined pre-knockout score ($r=-0.054$). Match-level accuracy produced a different ordering. Claude Sonnet 4.6 correctly predicted the largest number of group-stage outcomes (63.89\\%) but placed sixth overall. Average self-reported confidence was also unrelated to either outcome accuracy ($r=-0.060$) or total score ($r=-0.067$). The results suggest that forecasting a complete tournament tests something different from predicting matches one at a time, while also showing how strongly a bracket-based leaderboard can depend on scoring design. The benchmark materials, raw responses, and scoring code are released to support replication and future extensions." }, "2608.03528.pdf": { "id": "2608.03528v1", "title": "Tired Actor: Fatigue-Informed Character Control", "year": "2026", "authors": [ "Shengyuan Zhang", "Xinpeng Liu", "Muchun Niu", "Yulong Chen", "Lizhuang Ma", "Yue Gao" ], "categories": "cs.RO", "pdf_url": "https://arxiv.org/pdf/2608.03528v1", "abstract": "Replicating human behavior with physics simulation has been a long-expected goal in character animation. Existing efforts have achieved impressive performance in imitating a wide span of general motions. However, most existing efforts could still suffer from unnatural movements due to the lack of biomechanical and physiological priors. Given this, we project our sights to advances in behavioral energetics, which demonstrate how energy use shapes human movements. In contrast, current character controllers typically assume the character is equipped with infinite energy over time. Inspired by these, we propose to adopt fatigue as a proxy of the finite energy limit, inject it into general character animation, and thoroughly investigate how fatigue introduces new characteristics to physics-based character control. Leveraging the Three-Compartment Controller (3CC) model, we managed to obtain a policy for general motion imitation under different fatigue statuses. Furthermore, extensive analyses are conducted to demonstrate how fatigue could influence the naturalness, scalability, and robustness of character animation. Our code will be made public." }, "2608.05030.pdf": { "id": "2608.05030v1", "title": "From Score Matrices to Football-Aware Match-State Simulation: An Auditable LLM Harness for Exact-Score Reranking", "year": "2026", "authors": [ "Shaopeng Liang" ], "categories": "cs.AI", "pdf_url": "https://arxiv.org/pdf/2608.05030v1", "abstract": "Football score forecasting combines a strong statistical core with a difficult contextual edge. Dynamic Poisson-family models estimate team strength, expected goals, and coherent score probabilities, but do not directly understand roles, tactical matchups, motivation, or how a first goal changes behaviour. Large language models (LLMs) can reason about such concepts, yet are not calibrated probability engines. We combine both components through an auditable information harness. This paper documents four iterations: V1, a dynamic score-driven Dixon-Coles baseline; V2, which maps LLM contextual ratings back into expected-goal parameters; V3, which replaces scalar correction with goal-by-goal simulations over a frozen score-candidate set; and V4, which adds shared first-breakthrough and post-goal cascade judgments, time-aware stopping, and deterministic tail candidates. The harness defines input semantics, supplies pre-match evidence, and constrains the LLM to an inspectable reasoning route. On a chronological replay of the first 150 matches of the 2025-26 English Premier League, V1 achieved 10.0% Top-1 and 26.7% Top-3 exact-score accuracy. V3 reached 12.0% and 30.0%, while V4 reached 14.7% and 30.7%. V4 increased candidate coverage from 77.3% to 84.7%, although no added tail candidate became a Top-3 exact hit. V1's native 1X2 distribution achieved 53.3% argmax accuracy, 0.9878 log loss, 0.5870 Brier score, and 0.2095 ranked probability score. These results are exploratory: the development slice is not an untouched benchmark, and temporal input isolation cannot exclude outcome memory in a closed LLM. The contribution is an auditable hybrid architecture, a clear design evolution, and negative findings showing where football-aware simulation does and does not improve score selection." }, "2608.09887.pdf": { "id": "2608.09887v1", "title": "Space-Creating versus Dead Possession: An Off-Ball Possession-Quality Index for Broadcast Football", "year": "2026", "authors": [ "Seongjin Choi" ], "categories": "cs.CV, cs.CY, cs.LG", "pdf_url": "https://arxiv.org/pdf/2608.09887v1", "abstract": "Ball possession is the most-cited and most-misleading number in football: 60% recycled in one's own half is not 60% spent pinning the opponent back. Existing event-based possession-value frameworks (expected threat, VAEP, on-ball value) price on-ball actions but ignore the off-ball question a sterile possession poses: did holding the ball create space, or was the circulation dead? We answer this in two layers. First, an event-side junk-possession index prices each possession sequence by its peak threat gain under an expected-threat grid and -- after reconstructing the live scoreline to exclude lead-protecting circulation -- flags low-threat sequences in tied-or-losing states. On the 2026 FIFA World Cup (103 matches, 206 team-matches) the flag correlates negatively with points (r=-0.37) and xG difference (r=-0.51, partly index-coupled). It is not a repackaging of on-ball value: with team offensive VAEP and field tilt held fixed, the junk flag stays strongly negatively associated with points (p<0.0001, also match-clustered) while VAEP is not significant -- in this same-match (descriptive) regression it adds information beyond this on-ball action-value model. Second, for a flagged window we resolve whether it was spatially dead or space-creating by projecting broadcast video to pitch coordinates and measuring a Space-Creation Index (SCI): a net pitch-control change capturing whether the possession seized space or pushed the opponent's block back. Across 31 of 35 flagged windows from nine World Cup matches (a purposive sample), 74% are spatially non-space-creating, 19% weak progression, and 6% space-creating windows the event flag alone would score as failure -- including a side with 73% of the ball that exited on penalties (two non-creating windows). The two layers separate space-creating-but-unconverted from sterile possession, a distinction event-only on-ball value cannot make." }, "2608.11203.pdf": { "id": "2608.11203v1", "title": "Capturing Uncertainty in Human Motion for Representation Learning in Soccer", "year": "2026", "authors": [ "Yizhou Xu", "Lars Bretzner", "Tiesheng Wang", "Atsuto Maki" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2608.11203v1", "abstract": "This paper presents a self-supervised representation learning framework for understanding 3D skeleton-based human motion in soccer, using future motion prediction as the learning objective. Since human motion is inherently uncertain, accounting for multiple plausible futures is essential for capturing the underlying motion dynamics and learning effective representations. To this end, we introduce a conditioning module for motion prediction that models a probabilistic distribution over discretized future motions in 3D Euclidean space, learning multimodality with explicit supervision from future trajectories. Experiments on large-scale soccer player tracking data show that our approach substantially improves motion prediction accuracy. Moreover, the learned representations effectively transfer to multiple soccer downstream applications, demonstrating strong cross-task generalization." }, "2608.11505.pdf": { "id": "2608.11505v1", "title": "Does a Structural Model Add Anything to the Closing Price? Calibrated forecasting, incremental information, and match leverage in the Italian Serie A", "year": "2026", "authors": [ "Yannik Pitcan" ], "categories": "stat.AP, q-fin.ST", "pdf_url": "https://arxiv.org/pdf/2608.11505v1", "abstract": "Studies of association-football forecasting routinely report three-way accuracy in the low fifties and present it as competitive with the betting market. Accuracy against a uniform benchmark answers the wrong question; the question worth asking is whether a model carries information a margin-free closing price has not already absorbed. We formalise that test as the fitted weight in a logarithmic opinion pool and apply it to nineteen complete Serie A seasons (7,220 matches). The answer is negative and stable. A Dixon-Coles model with tuned exponential decay attains 53.4% accuracy and a Ranked Probability Score of 0.1972 against the market's 0.1905; the paired difference is +0.0067 (95% CI [0.0046, 0.0088]) and the market wins in all seven test seasons. The fitted pooling weight on the structural model is 0.000, and the log-loss profile is monotone increasing in that weight on validation and test alike, so this is a boundary solution, not an optimisation artefact. Refitting the same machinery to shots on target yields a variant earning weight 0.35 against the goals model -- it carries information the goals model lacks -- and 0.000 against the market. Two structural signals, each informative about the other, both priced. The structural model is better calibrated than the market on the home-win margin (slope 0.995 versus 1.103) while clearly less sharp: the market's advantage is discrimination rather than honesty, which accuracy alone cannot distinguish. Value lies not in a better forecast but in what is built on a calibrated one. We define match leverage, the change in a club's probability of achieving a season objective between winning and losing a fixture, and compute it for ACF Fiorentina: an away fixture against a relegation rival carried 2.25x the leverage of hosting the eventual champions. The paper also documents and corrects errors in an earlier study of our own." }, "2608.12926.pdf": { "id": "2608.12926v1", "title": "H-VAEP and H-xT: Valuing Offensive On-the-Ball Actions in Handball by Estimating Probabilities", "year": "2026", "authors": [ "Julius Broermann", "Oliver M\u00fcller", "Michael D\u00f6ring", "Jochen Baumeister" ], "categories": "cs.LG, cs.AI", "pdf_url": "https://arxiv.org/pdf/2608.12926v1", "abstract": "Traditional player evaluation in professional handball relies on basic box-score metrics or heuristic indices, which fail to credit the multi-player build-up chain. While football (soccer) analytics has adopted Expected Threat (xT) and Valuing Actions by Estimating Probabilities (VAEP), these event-based action valuation frameworks have not yet been adapted to handball. In this paper, we present the first comprehensive adaptation and evaluation of xT and VAEP for handball, utilizing five seasons of tracking-derived event data from the Handball Bundesliga. We develop Handball-xT (H-xT) using a handball-native court zoning layout, demonstrating via simulations that it is systematically more robust than standard rectangular grids. We optimize Handball-VAEP (H-VAEP) by tailoring its feature space and selecting the context length to limit team-identity leakage. Our evaluation shows that H-VAEP yields exceptionally stable, discriminative, and intuitive player ratings that highlight build-up play. Finally, we release our complete code repository to help professional clubs deploy these models." }, "2608.19723.pdf": { "id": "2608.19723v1", "title": "StreamSoccer: Event-Driven Memory for Streaming Soccer Commentary", "year": "2026", "authors": [ "Chenxi Shao", "Bozhong Wang", "Jiaxin Huang", "Zhao Liu", "Sunwei Zhu", "Tianxin Hang" ], "categories": "cs.CV, cs.CL", "pdf_url": "https://arxiv.org/pdf/2608.19723v1", "abstract": "Streaming video understanding requires models to causally update state as video arrives and organize growing history into semantic units that can evolve, persist, and be recalled under bounded computation and memory. This challenge is pronounced in live soccer commentary, where a system must describe completed events, summarize recent play, recall earlier events, or remain silent using only information available before each utterance. We present StreamSoccer, an event-driven system that uses event memory as its intermediate representation. A fixed-budget active memory integrates the stream; completed event states are retained locally and consolidated into retrievable historical records. A unified generator uses current, recent, and historical context to produce three commentary modes, while a rule-assisted scheduler selects a mode or silence. Unlike streaming video-language models organized around frames, visual tokens, or caches, and soccer-commentary methods based on predefined clips or output timestamps, StreamSoccer explicitly models event lifecycles. We construct a three-track streaming soccer commentary dataset and a layered evaluation protocol. At common reference anchors, StreamSoccer obtains CIDEr scores of 38.62, 23.96, and 17.39 for current-event, recent-window, and historical-memory commentary, ranking first on the current-event and historical-memory tracks and second on recent-window. Controlled ablations show that local completed events improve all tracks and that the full system performs best on all three. Across 174 raw-video runs on 58 matches, per-minute RTF p95 ranges from 0.10 to 0.22 without sustained growth with match history. These results indicate that event memory supports streaming soccer commentary across temporal scopes while controlling long-history computation." }, "2608.20395.pdf": { "id": "2608.20395v2", "title": "The Effective Gravitational Field of the Ball in Association Football", "year": "2026", "authors": [ "Oliver B. Wright" ], "categories": "physics.soc-ph", "pdf_url": "https://arxiv.org/pdf/2608.20395v2", "abstract": "In association football, collective player motion is organized around the ball. We ask whether this many-agent motion can be described by an effective field analogous to gravitational attraction, emphasizing that the measured quantity is a radial drift velocity, as in overdamped dynamics, rather than a Newtonian acceleration. Using public tracking data from ten matches, we characterize this radial drift velocity through the distance-dependent mean D(r) and the ball-position-dependent field K(x,y). After excluding restarts and other constrained phases of play, both goalkeepers, the player nearest to the ball, and small player-ball separations, a persistent inward drift velocity remains, with a global mean of about 0.9 m/s. The distance dependence is not described by an inverse power of r: D(r) decreases at short range and then forms a broad plateau that persists to the largest measured separations. The mean K(x,y) is positive over most of the pitch, with a central depression about 15 percent below the non-goalmouth average and stronger suppression near the goalmouths, where the mean drift velocity becomes slightly negative, corresponding to weak effective repulsion. The role-resolved D(r) shows a pronounced defender minimum near 12 m and a pronounced attacker maximum near 60 m, whereas only defenders produce effective repulsion near the goals in K(x,y). These results show that football tracking data can reveal simple effective laws of active many-body motion while retaining clear signatures of player role and pitch geometry." }, "2608.21530.pdf": { "id": "2608.21530v1", "title": "Multimodal Injury Risk and Performance Prediction in Tennis Using Weighted Ensemble Learning", "year": "2026", "authors": [ "Weihao Qu", "Dongyang Wang", "Ling Zheng", "Francisco E. Alvarez", "Shobharani Polasa", "Jiacun Wang" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2608.21530v1", "abstract": "Machine learning has had a positive impact on the sports industry, with one of its most promising applications being the prediction of athlete performance and injury risk. Recent advances have employed state-of-the-art models to improve prediction accuracy, yet progress remains limited by data availability and the reliance on subjective observations or expert assessments. To address these limitations, researchers in sports such as soccer, basketball, and wrestling have begun integrating heterogeneous data sources, such as wearable device readings, with traditional subjective assessments. However, similar multimodal approaches remain underexplored in tennis. In this work, we propose a multimodal weighted ensemble learning framework, Predictive Athlete Readiness for Tennis (PART), to monitor athlete wellness and estimate near-term injury risk in tennis players. PART processes a wide range of inputs, including physiological metrics, training and match data, sleep information from wearable devices, self-reported questionnaires, vertical jump assessments, and motion analysis from match-play videos. From these modalities, specialized machine learning and deep learning models independently extract four athlete-specific characteristics: overall wellness, injury risk, physical capability, and playing style. To overcome the complexity of combining these diverse modalities, PART employs a supervised weighted ensemble integration strategy, assigning adaptive weights to each predictive model based on its reliability. Evaluation of multimodal data collected from nine collegiate tennis players demonstrates that PART achieves strong performance in monitoring athlete wellness and estimating near-term injury susceptibility. Beyond collegiate athletes, the framework also shows promise for recreational tennis players, offering personalized insights to mitigate injury risk and optimize performance." }, "2608.25126.pdf": { "id": "2608.25126v1", "title": "Multimodal Injury Risk Prediction in Tennis", "year": "2026", "authors": [ "Francisco Erramuspe Alvarez", "Shobharani Polasa", "Weihao Qu", "Jay Wang", "Ling Zheng" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2608.25126v1", "abstract": "Machine learning has had a significant positive impact on the prediction of athlete performance and injury risk. Most works in this field rely on subjective observations and expert assessments, which restrict their effectiveness. In sports like soccer, basketball, and wrestling, some studies attempt to address this challenge by integrating data from alternative sources, such as readings from wearable devices, alongside traditional subjective observations and expert assessments to enhance accuracy. However, similar research in tennis remains largely unexplored. In this paper, we propose a multimodal Predictive Athlete Readiness framework for Tennis (PART) to assess both performance and injury risk in tennis players. By leveraging machine learning and deep learning techniques, PART processes multiple sources of data collected from nine collegiate tennis players, including physiological metrics, training and match data, sleep data from wearable devices, self-reported information via daily questionnaires, jump assessments, and motion analysis from match play videos. PART captures four characteristics of tennis players: overall wellness, injury risk, physical capability, and playing style. By integrating these four characteristics by supervised learning, it is capable of providing a holistic assessment of the tennis athlete's condition, along with advanced forecasts of specific body areas at risk such as the upper body (e.g., elbows) or lower body (e.g., knees). Our evaluation, conducted with data from nine collegiate tennis players, shows that PART achieves strong performance in predicting both overall wellness and injury risk. Additionally, our framework also shows promise for recreational tennis players, who often suffer from injuries due to incorrect playing techniques." }, "2608.25962.pdf": { "id": "2608.25962v1", "title": "Simulating an imaging atmospheric radio telescope to observe cosmic gamma rays and cosmic neutrinos", "year": "2026", "authors": [ "Sebastian Achim Mueller", "Anne Timmermans", "Juan Ammerman-Yebra", "Harm Schoorlemmer" ], "categories": "astro-ph.IM", "pdf_url": "https://arxiv.org/pdf/2608.25962v1", "abstract": "Observations of cosmic gamma~rays with energies above the 10$^{12}$ electronvolts (TeV) regime are often a direct probe of events where our theoretical predictions are challenged by extreme conditions. Because high energetic cosmic gamma-rays are rare, one needs collection areas with the size of soccer fields (10$^{5}$ m$^2$) or more to gather significant counting statistics in short time. The imaging atmospheric Cherenkov telescope detects cosmic gamma-rays in such large areas and currently offers the best reconstruction power for the cosmic particle's type, energy, and direction in particular. However, with only about 1,000 hours of dark and clear nights every year, the imaging atmospheric Cherenkov~telescope only has a duty cycle of about $12\\%$. To overcome this limitation, we propose a novel imaging atmospheric radio telescope. By observing radio emission in the $10$ GHz regime from air showers, the imaging atmospheric radio telescope could observe 24/7, independent of the day night cycle, and almost independent of the weather. The novelty here is the high resolution imaging of the air shower's radio emission which might result in high resolution images as one finds them in imaging atmospheric Cherenkov~telescopes. We present a technique to simulate the image formation in the radio telescope using wave mechanics instead of conventional ray tracing. The imaging atmospheric radio telescope could increase our access more than eight-fold to an accurately reconstructed high energy gamma-ray-sky. Further, the imaging atmospheric radio telescope might also eight fold the observation power for cosmic neutrinos in the so called Earth-skimming technique." }, "2608.29487.pdf": { "id": "2608.29487v1", "title": "Blind Dexterity: Whole-Body Humanoid Manipulation via Pure Proprioception", "year": "2026", "authors": [ "Aditya Bhatt", "Oleg Kaidanov", "Puze Liu", "Jan Peters" ], "categories": "cs.RO", "pdf_url": "https://arxiv.org/pdf/2608.29487v1", "abstract": "We present blind, whole-body manipulation skills on a Unitree G1 humanoid using only onboard proprioception, without cameras, markers, force-torque, or tactile sensors. Despite this minimal sensing, the trained policies exhibit surprising capability across qualitatively different tasks: push-resilient bipedal walking without IMU feedback, active soccer ball trapping with a foot, seeking and lifting a suitcase by its handle, and mounting a randomly positioned skateboard. We argue that these capabilities arise from a key underappreciated signal: the way the joint encoder readouts evolve under purposeful compliant contact, effectively forming a whole-body tactile channel. By generating contact-rich motions, the trained policies actively probe the environment; as a result, task-relevant object state (e.g., pose) becomes increasingly decodable from short proprioceptive histories. We expose this information using compact task-specific state estimators trained alongside, but fully separately from, the policies; their prediction errors decrease rapidly after informative contact. Our results indicate that joint encoder-based proprioception, combined with compliant actuation (now widely available on commercial robots and low-cost motors) is already a strong, practical substrate for whole-body dexterous manipulation and interactive perception, and therefore a natural foundation on which richer sensing can be layered." }, "2608.30227.pdf": { "id": "2608.30227v1", "title": "ELASTIC: Trajectory-Based Synchronization of Event and Tracking Data in Soccer", "year": "2026", "authors": [ "Hyunsung Kim", "Hoyoung Choi", "Kunhee Lee", "Sangwoo Seo", "Tom Boomstra", "Jinsung Yoon" ], "categories": "cs.DB", "pdf_url": "https://arxiv.org/pdf/2608.30227v1", "abstract": "Combining event and tracking data is fundamental to modern soccer analytics, yet the two sources are rarely well aligned: event timestamps recorded by human annotators often miss the true moment of the action, distorting the spatiotemporal context that downstream models rely on. Existing synchronization methods depend on noisy human-annotated event locations and fail to detect ball receptions, obscuring when each player gains ball possession. To address these limitations, we propose ELASTIC (Event-Location-AgnoSTIC synchronizer), a framework that infers the start and end timestamps of events solely from player and ball trajectories, without relying on annotated event locations. To recover ball receptions, ELASTIC enriches the event sequence by inserting virtual termination events between consecutive events, so that the end of each event is detected jointly with its start. It then extracts a sparse set of candidate frames where ball touches are physically plausible, and aligns the termination-inserted event sequence with the candidate-frame sequence using an extended Needleman-Wunsch algorithm. For reproducible evaluation, we construct a publicly available benchmark by annotating ground-truth timestamps on the Sportec Open DFL Dataset, on which ELASTIC substantially outperforms existing methods. Through downstream task evaluation, we further show that improved synchronization translates into measurable gains in soccer analytics. The source code and benchmark are available at https://github.com/hyunsungkim-ds/elastic.git." }, "2609.03790.pdf": { "id": "2609.03790v2", "title": "When Minute-Resolution Monitoring Meets Session-Level Injury Labels: Landmark-Based Discrimination in Elite Women's Football", "year": "2026", "authors": [ "Evangelos Chatzidimitriou", "Konstantinos Tserpes" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2609.03790v2", "abstract": "Minute-resolution athlete monitoring is increasingly common, while injury annotation may exist only at the athlete-session level and omit within-session onset time. Replicating a positive session label across every recorded minute would therefore create unsupported minute-level supervision. We address this label-resolution mismatch using fixed elapsed-time landmarks at 10, 20, 30, 40, 50, and 60 min, constructing one representation per athlete-session from information available up to each landmark while keeping the target as a same-day injury-associated session indicator. Using 2020 SoccerMon data from elite women's football, the modelling cohort contains 2,259 Team A athlete-sessions from 27 athletes, including all 22 positive sessions from five athletes. Evaluation is athlete-disjoint. We compare contextual, cumulative, and dynamic representations; Logistic Regression, Random Forest, XGBoost, and TabPFN; and training-only NONE, SMOTE, and CTGAN conditions. Robustness is assessed using athlete-cluster bootstrap, a fixed common cohort, alternative fold allocations, leave-one-positive-athlete-out analysis, and equal-athlete weighting. Discrimination is landmark-dependent and non-monotonic. TabPFN improves later-landmark discrimination relative to Logistic Regression but does not consistently outperform Random Forest. Synthetic augmentation provides condition-specific rather than universal benefit. The contribution is a unit-aligned framework for using minute-resolution predictors with session-level supervision, not minute-specific injury prediction." }, "2609.04803.pdf": { "id": "2609.04803v1", "title": "Hierarchical Possession-Aware Graph Pointer Network for Pass Receiver Selection", "year": "2026", "authors": [ "Jingyi Wang", "Da Li", "Kaixin Wang", "Zhangqin Huang" ], "categories": "cs.AI, cs.SI", "pdf_url": "https://arxiv.org/pdf/2609.04803v1", "abstract": "Pass receiver selection is a fundamental task in football analytics, aiming to predict the intended receiver under a given game state. This task is challenging with event-centered freeze-frame observations, a broadcast-like setting that provides only partial and variable player visibility without complete trajectories or stable player identities. The model must therefore reason over anonymous visible candidates, opponent pressure, and recent context under partial observation. To address this setting, we propose a Hierarchical Possession-aware Graph Pointer Network (HPGPN), which formulates pass receiver selection as variable-size candidate prediction over visible teammates. HPGPN jointly models current player interactions, local event context, and possession-level temporal dynamics. It represents the current pass situation with a graph, incorporates fixed event context, and uses dynamic possession history to capture how the attacking sequence evolves. Candidate representations are refined hierarchically by integrating spatial, contextual, and historical evidence, and a glimpse pointer head scores the receiver candidates. Experiments on public football event and freeze-frame data show that HPGPN improves pass receiver selection performance. Ablation studies demonstrate the effectiveness of graph-based interaction modeling, fixed event context, and dual-branch dynamic possession-history modeling." }, "2609.06610.pdf": { "id": "2609.06610v1", "title": "A Statistical and Machine Learning Framework for Quantifying Offensive Impact in Professional Box Lacrosse", "year": "2026", "authors": [ "Robert Jimerson" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2609.06610v1", "abstract": "Professional box-lacrosse statistics summarize outcomes but provide limited information about shot quality or the roles behind scoring opportunities. This study develops a documented framework for estimating expected goals (xG) and attributing recorded offensive involvement using 1,006 manually annotated Rochester Knighthawks shot attempts, including 151 goals, from 13 consecutive 2025-2026 National Lacrosse League games. Logistic regression, random forest, and extremely randomized trees were evaluated across three nested feature sets using Leave-One-Game-Out cross-validation and a training-fold base-rate benchmark. The contextual baseline random forest had the lowest observed pooled log loss (0.4189) and Brier score (0.1260), improving on the benchmark by 1.22% and 1.50%; five of nine specifications did not beat the benchmark. Adding two-man-action and pick-type fields did not improve the primary metrics. Core Offensive Impact attributes recorded involvement through shooter xG and shot-based expected assists for final passers. Expected Pick Value (xPV) compares a qualifying pick's observed-state probability with a no-pick counterfactual. Its magnitude was indistinguishable from model noise. Its directional pattern exceeded 200 row-permutation replicates, but limited tail resolution and failure to preserve game-level pick composition make the diagnostic descriptive rather than inferential. Accordingly, xPV is reported only as an exploratory augmented component. Given the single-team, 13-game sample, the results are an initial case study rather than league-wide or causal estimates." }, "2609.06718.pdf": { "id": "2609.06718v2", "title": "SkillX: Unified Multi-Skill Policy Learning for Humanoid Soccer", "year": "2026", "authors": [ "Zhangchen Ye", "Enxuan Ruan", "Yifei Bao", "Runhan Huang", "Jiankun Yang", "Jiakang Jin" ], "categories": "cs.RO", "pdf_url": "https://arxiv.org/pdf/2609.06718v2", "abstract": "Humanoid soccer is a challenging testbed for dynamic whole-body control, requiring robots to coordinate balance, locomotion, object interaction, and skill switching over long horizons. Existing humanoid sports methods often rely on task-specific multi-stage pipelines, making it difficult to jointly learn and compose multiple object-interactive skills within a single deployable policy. To address this, we present SkillX, a unified reinforcement learning framework that learns and composes multiple atomic soccer skills through a single command-conditioned policy. SkillX integrates three core designs: skill-specific adversarial motion priors, skill-specific critics, and an object-aware temporal encoder, enabling the robot to execute atomic skills and transition among them such as dribbling, trapping, and shooting. Experiments in simulation and on a real Noetix E1 humanoid demonstrate robust multi-skill execution, long-horizon skill composition, and successful sim-to-real deployment." }, "2609.10498.pdf": { "id": "2609.10498v1", "title": "Field Converter: Geometry-Initialized Temporal Residual Refinement for World-Grounded Player Pose Estimation from Soccer Broadcasts", "year": "2026", "authors": [ "Simon Khan", "Laurent Gajny", "Jennyfer Lecompte", "S\u00e9bastien Laporte" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2609.10498v1", "abstract": "Recovering 3D human pose from monocular sports broadcasts remains challenging when players must be localized in a shared metric world coordinate system rather than only reconstructed relative to their own body. We introduce Field Converter, a geometry-initialized temporal residual framework for world-grounded 3D player pose estimation from calibrated soccer broadcasts. Our method first uses camera and pitch geometry to initialize the player root through ray-ground intersection, then predicts a temporal residual correction from pose, image, camera, and geometric cues. On match-disjoint evaluation sequences, residual refinement reduces root error from 49cm with geometry alone to 14cm with a frame-wise MLP and 10cm with a TCN, while a Transformer achieves a comparable 11cm. The resulting world-space MPJPE reaches 13.2cm, and ablations show that residual prediction clearly outperforms direct global-root regression while temporal context matters more than the specific temporal backbone. Failure analysis further identifies airborne motion as the main limitation of the ground-based geometric initialization." }, "2609.24040.pdf": { "id": "2609.24040v1", "title": "Graph-to-Grid (G2G): Continuous-Coordinate Feature Painting for Soccer Pass Surfaces", "year": "2026", "authors": [ "Kaan G\u00fcnay", "Orhun Gun" ], "categories": "cs.LG, cs.CV", "pdf_url": "https://arxiv.org/pdf/2609.24040v1", "abstract": "Dense pass surfaces give, for every pitch cell, whether a pass played there would arrive, whether the carrier would choose it, and what the possession would then be worth. The networks that draw them read the state as a raster of per-cell counts, losing where inside a cell each player stands. LiDAR detectors, bird's-eye-view perception and graph weather models move entity features onto a grid, binning each entity to a cell or learning the transfer. We evaluate the interpolated form: each player's features are scattered bilinearly onto the grid at the player's measured coordinates, so the surface loss trains the per-player encoder end to end. Those systems adopt an interface; this paper measures one. On 53,628 passes from the 2022 World Cup, painting improves selection likelihood over the same core fed rasters alone by about a quarter of a nat: in every match of an eight-fold cross-validation, with every arm tuned over five seeds, and after retraining on seven Bundesliga and 2. Bundesliga matches from another provider. Thirteen pre-specified studies locate the gain: painting the nine raw player features with no encoder carries three quarters of it, and the learned encoder and message passing add a smaller, resolved increment. Painting also helps the original SoccerMap and a canonical U-Net, whereas offset channels, a finer raster, an attention painter and a raster-free decoder do not. Frozen across the provider boundary the likelihood advantage is lost; injected tracking error compresses it. These results concern observed-endpoint prediction, not calibrated evaluation of hypothetical passes." }, "2609.25569.pdf": { "id": "2609.25569v1", "title": "SambaGraph: Action-Reaction Spatio-Temporal Graphs for Soccer Tactical Response Modeling", "year": "2026", "authors": [ "Abel A. Reyes-Angulo", "Henry O. Velesaca", "Steven Araujo" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2609.25569v1", "abstract": "Soccer tactics are interactive: an attacking action changes the opponent's defensive problem, and the observed response depends on the multi-agent match state. We introduce SambaGraph, an action--reaction spatio-temporal graph dataset and benchmark for soccer tactical response modeling. From tracking and event data for all 64 matches of the 2022 FIFA World Cup, we curate 4,070 action-centered episodes represented as temporally aligned 23-node player--ball graph sequences with attack/defense views, response labels, and 26,270 split-safe attack--defense pairs. We study three questions: whether observed responses can be classified from graph episodes, whether successful defenses can be retrieved for a query attack, and whether graph-derived summaries support grounded LLM reasoning. A compact signature MLP obtains $0.796\\pm0.007$ macro-F1 for response classification, while a fused graph--signature dual encoder reaches $0.471\\pm0.029$ Hit@5 and $0.655\\pm0.051$ Hit@10 for full-bank defensive retrieval. Hard negatives maximize pair discrimination but not retrieval quality. Local LLMs underperform supervised encoders for direct classification and do not improve over a strong original order in eight-candidate reranking, but they provide grounded tactical rationales. These results position SambaGraph as a reproducible benchmark for graph-based soccer strategy-response research. Code and dataset are available at: https://github.com/areyesan/SambaGraph." }, "2609.28175.pdf": { "id": "2609.28175v1", "title": "DAVIS: A Depth-Only End-to-End Active-Vision Framework for Humanoid Soccer Skills", "year": "2026", "authors": [ "Jiakang Jin", "Yixiao Huo", "Pengyuan Wang", "Yinan Han", "Tingxuan Zhang", "Zhuobing Zhao" ], "categories": "cs.RO", "pdf_url": "https://arxiv.org/pdf/2609.28175v1", "abstract": "Humanoid soccer contact skills require more than producing high-impact foot-ball contacts: the robot must close the loop over perception, approach, alignment, impact, and recovery while its own motion induces substantial viewpoint changes, frequent loss of the ball from view, and uncertain contact outcomes. In this work, we ask a compact yet stricter question: can a humanoid learn soccer contact skills using only a head-mounted depth image, proprioceptive history, and an optional low-dimensional task command, and directly output 25-DoF joint PD targets without extra runtime perception or planning modules? To this end, we propose DAVIS, a depth-only end-to-end framework for humanoid soccer skills that learns visibility-aware auxiliary geometry during training, and combines GT-to-prediction annealing, task curricula, and AMP-style motion priors to smoothly bridge privileged supervision and real deployment. Built on this framework, we instantiate representative soccer contact skills, including goal-directed shooting and directional dribbling, through task-specific definitions of objects, commands, rewards, and curricula, and validate them through simulation, Noetix E1 real-robot experiments, and ablations." }, "2609.38588.pdf": { "id": "2609.38588v1", "title": "Drone Soccer: Learning to Manipulate with Multicopter Downwash", "year": "2026", "authors": [ "Neelay Joglekar", "Bavin Saravanan", "Yutong Wang", "Varun Kandiyappan", "Junyi Geng", "Sebastian Scherer" ], "categories": "cs.RO", "pdf_url": "https://arxiv.org/pdf/2609.38588v1", "abstract": "Although multicopter drones are traditionally designed for \"perception-only\" tasks, like mapping and exploration, recent work has sought to develop Unmanned Aerial Manipulators (UAMs) to solve mobile manipulation tasks. Aerial manipulation performance can be impacted by \"downwash,\" the airflow produced by propellers, but current state-of-the-art UAMs either ignore downwash or treat it as a disturbance. Instead, is it possible to actively use downwash as a tool during manipulation? We design a drone soccer task to explore the feasibility of downwash-based manipulation. Specifically, we develop a simplified downwash dynamics model which we use to train an RL policy to dribble a soccer ball. We further demonstrate that our policy transfers to real world deployment. This work provides key insights into novel manipulation capabilities for multicopters." }, "2609.39300.pdf": { "id": "2609.39300v1", "title": "BMASH: Ball-Motion-Aware Soccer Header Spotting", "year": "2026", "authors": [ "Ahmed Endris Hasen", "Muhammad Shahzad Khan", "Nikolaos Passalis", "Jenni Raitoharju" ], "categories": "cs.CV", "pdf_url": "https://arxiv.org/pdf/2609.39300v1", "abstract": "Recent advances in computer vision have made broadcast sports videos increasingly useful for event analysis, performance assessment, and player-safety applications. In soccer, however, header spotting remains a challenging problem due to the subtle and short-lived nature of header events. This paper focuses on soccer header spotting: identifying moments in broadcast videos where the ball contacts a player's head. We first adapt and evaluate Video Swin as a strong action-recognition baseline for this task, and then introduce BMASH, a ball-motion-aware fusion framework that integrates detector-derived ball features. BMASH combines Video Swin action representations with ball-presence and motion features from frame-level soccer-ball detection, integrating player-action context with ball dynamics to distinguish headers from visually similar events. We evaluate BMASH using game-level splits with separate test matches and rotating validation folds, considering both centered-window classification and continuous full-video spotting. Results show that Video Swin provides a strong baseline for header spotting, while BMASH improves clip-level AP and ROC-AUC over the corresponding Video Swin baseline. In continuous full-video spotting, BMASH achieves a comparable event-level F1-performance with a different precision--recall trade-off." }, "2610.05961.pdf": { "id": "2610.05961v1", "title": "Strategic Multi-Agent Learning for Interpretable Action Valuation of All Players in Football", "year": "2026", "authors": [ "Kenjiro Ide", "Taiga Someya", "Kohei Kawaguchi", "Keisuke Fujii" ], "categories": "cs.LG", "pdf_url": "https://arxiv.org/pdf/2610.05961v1", "abstract": "Valuing player actions in football requires accounting for strategic interactions among 22 players, including off-ball movements and defensive positioning. Existing reinforcement-learning-based methods commonly aggregate decisions at the team level or estimate player values independently, leaving strategic interdependence among players insufficiently represented. This study proposes an action valuation framework inspired by Markov perfect equilibrium (MPE) for all players. Each possession is modeled as a finite-horizon dynamic game, with each player represented as an autonomous agent whose policy depends on the current game state. MPE is used as a motivating solution concept rather than an exact equilibrium. To improve interpretability, we use Expandable Decision-Making States (EDMS) and decompose the Q-value into a successor-feature basis and a linear reward-weight vector. The value basis is estimated by linear TD initialization followed by nonlinear refinement. Using tracking and event data from 95 J1 League matches, we compare the proposed formulation with an independent reinforcement learning baseline. Because the two formulations define TD errors in different target spaces, TD MSE is used only for within-formulation consistency. With EDMS fixed, the independent baseline assigns the highest value to forward movement in 99.21% of evaluated off-ball states, whereas the most frequent direction under the proposed formulation accounts for 17.63%. Team-level average Q-values show a negative association with season-level expected goals for the baseline and a weakly positive association for the proposed formulation. Qualitative analyses illustrate context-dependent valuations of off-ball movements and defensive positioning. Overall, the proposed formulation produces more context-sensitive action rankings, although the comparison does not isolate the MPE-inspired component." }, "2506.16696.pdf": { "id": "2506.16696v1", "title": "Interpretable Low-Dimensional Modeling of Spatiotemporal Agent States for Decision Making in Football Tactics", "year": "2025", "authors": [ "Kenjiro Ide", "Taiga Someya", "Kohei Kawaguchi", "Keisuke Fujii" ], "categories": [ "cs.AI" ], "pdf_url": "https://arxiv.org/pdf/2506.16696v1", "abstract": "Understanding football tactics is crucial for managers and analysts. Previous research has proposed models based on spatial and kinematic equations, but these are computationally expensive. Also, Reinforcement learning approaches use player positions and velocities but lack interpretability and require large datasets. Rule-based models align with expert knowledge but have not fully considered all players' states. This study explores whether low-dimensional, rule-based models using spatiotemporal data can effectively capture football tactics. Our approach defines interpretable state variables for both the ball-holder and potential pass receivers, based on criteria that explore options like passing. Through discussions with a manager, we identified key variables representing the game state. We then used StatsBomb event data and SkillCorner tracking data from the 2023$/$24 LaLiga season to train an XGBoost model to predict pass success. The analysis revealed that the distance between the player and the ball, as well as the player's space score, were key factors in determining successful passes. Our interpretable low-dimensional modeling facilitates tactical analysis through the use of intuitive variables and provides practical value as a tool to support decision-making in football." }, "2505.01902.pdf": { "id": "2505.01902v1", "title": "From Players to Champions: A Generalizable Machine Learning Approach for Match Outcome Prediction with Insights from the FIFA World Cup", "year": "2025", "authors": [ "Ali Al-Bustami", "Zaid Ghazal" ], "categories": [ "cs.LG" ], "pdf_url": "https://arxiv.org/pdf/2505.01902v1", "abstract": "Accurate prediction of FIFA World Cup match outcomes holds significant value for analysts, coaches, bettors, and fans. This paper presents a machine learning framework specifically designed to forecast match winners in FIFA World Cup. By integrating both team-level historical data and player-specific performance metrics such as goals, assists, passing accuracy, and tackles, we capture nuanced interactions often overlooked by traditional aggregate models. Our methodology processes multi-year data to create year-specific team profiles that account for evolving rosters and player development. We employ classification techniques complemented by dimensionality reduction and hyperparameter optimization, to yield robust predictive models. Experimental results on data from the FIFA 2022 World Cup demonstrate our approach's superior accuracy compared to baseline method. Our findings highlight the importance of incorporating individual player attributes and team-level composition to enhance predictive performance, offering new insights into player synergy, strategic match-ups, and tournament progression scenarios. This work underscores the transformative potential of rich, player-centric data in sports analytics, setting a foundation for future exploration of advanced learning architectures such as graph neural networks to model complex team interactions." }, "2604.11786.pdf": { "id": "2604.11786v1", "title": "GenTac: Generative Modeling and Forecasting of Soccer Tactics", "year": "2026", "authors": [ "Jiayuan Rao", "Tianlin Gui", "Haoning Wu", "Yanfeng Wang", "Weidi Xie" ], "categories": [ "cs.AI", "cs.MA" ], "pdf_url": "https://arxiv.org/pdf/2604.11786v1", "abstract": "Modeling open-play soccer tactics is a formidable challenge due to the stochastic, multi-agent nature of the game. Existing computational approaches typically produce single, deterministic trajectory forecasts or focus on highly structured set-pieces, fundamentally failing to capture the inherent variance and branching possibilities of real-world match evolution. Here, we introduce GenTac, a diffusion-based generative framework that conceptualizes soccer tactics as a stochastic process over continuous multi-player trajectories and discrete semantic events. By learning the underlying distribution of player movements from historical tracking data, GenTac samples diverse, plausible, long-horizon future trajectories. The framework supports rich contextual conditioning, including opponent behavior, specific team or league playing styles, and strategic objectives, while grounding continuous spatial dynamics into a 15-class tactical event space. Extensive evaluations on our proposed benchmark, TacBench, demonstrate four key capabilities: (1) GenTac achieves high geometric accuracy while strictly preserving the collective structural consistency of the team; (2) it accurately simulates stylistic nuances, distinguishing between specific teams (e.g., Auckland FC) and leagues (e.g., A-League versus German leagues); (3) it enables controllable counterfactual simulations, demonstrably altering spatial control and expected threat metrics based on offensive or defensive guidance; and (4) it reliably anticipates future tactical outcomes directly from generated rollouts. Finally, we demonstrate that GenTac can be successfully trained to generalize to other dynamic team sports, including basketball, American football, and ice hockey." }, "2509.04731.pdf": { "id": "2509.04731v3", "title": "Generative World Models of Tasks: LLM-Driven Hierarchical Scaffolding for Embodied Agents", "year": "2025", "authors": [ "Brennen Hill" ], "categories": [ "cs.AI", "cs.CL", "cs.LG", "cs.MA", "cs.RO" ], "pdf_url": "https://arxiv.org/pdf/2509.04731v3", "abstract": "Recent advances in agent development have focused on scaling model size and raw interaction data, mirroring successes in large language models. However, for complex, long-horizon multi-agent tasks such as robotic soccer, this end-to-end approach often fails due to intractable exploration spaces and sparse rewards. We propose that an effective world model for decision-making must model the world's physics and also its task semantics. A systematic review of 2024 research in low-resource multi-agent soccer reveals a clear trend towards integrating symbolic and hierarchical methods, such as Hierarchical Task Networks (HTNs) and Bayesian Strategy Networks (BSNs), with multi-agent reinforcement learning (MARL). These methods decompose complex goals into manageable subgoals, creating an intrinsic curriculum that shapes agent learning. We formalize this trend into a framework for Hierarchical Task Environments (HTEs), which are essential for bridging the gap between simple, reactive behaviors and sophisticated, strategic team play. Our framework incorporates the use of Large Language Models (LLMs) as generative world models of tasks, capable of dynamically generating this scaffolding. We argue that HTEs provide a mechanism to guide exploration, generate meaningful learning signals, and train agents to internalize hierarchical structure, enabling the development of more capable and general-purpose agents with greater sample efficiency than purely end-to-end approaches." }, "2307.13807.pdf": { "id": "2307.13807v1", "title": "Sports Betting: an application of neural networks and modern portfolio theory to the English Premier League", "year": "2023", "authors": [ "V\u00e9lez Jim\u00e9nez", "Rom\u00e1n Alberto", "Lecuanda Ontiveros", "Jos\u00e9 Manuel", "Edgar Possani" ], "categories": [ "q-fin.PM", "cs.LG" ], "pdf_url": "https://arxiv.org/pdf/2307.13807v1", "abstract": "This paper presents a novel approach for optimizing betting strategies in sports gambling by integrating Von Neumann-Morgenstern Expected Utility Theory, deep learning techniques, and advanced formulations of the Kelly Criterion. By combining neural network models with portfolio optimization, our method achieved remarkable profits of 135.8% relative to the initial wealth during the latter half of the 20/21 season of the English Premier League. We explore complete and restricted strategies, evaluating their performance, risk management, and diversification. A deep neural network model is developed to forecast match outcomes, addressing challenges such as limited variables. Our research provides valuable insights and practical applications in the field of sports betting and predictive modeling." }, "2603.21020.pdf": { "id": "2603.21020v1", "title": "Characterizing Long-Range Dependencies in Knee Joint Contact Mechanics: A Comparison of Topology Diffusion, Global Routing, and Hybrid Graph Neural Networks", "year": "2026", "authors": [ "Zhengye Pan", "Jianwei Zuo", "Jiajia Luo" ], "categories": [ "q-bio.QM" ], "pdf_url": "https://arxiv.org/pdf/2603.21020v1", "abstract": "Finite element analysis of knee joint contact mechanics is computationally expensive, which has motivated the development of graph neural network surrogate models. However, effectively representing long-range dependencies in joint mechanical responses remains challenging. This study systematically compared topology diffusion, global routing, and their hybridization for surrogate modeling of knee joint contact mechanics. Using kinematic and force data from nine soccer players performing change-of-direction maneuvers, finite element simulations were used to generate graph-structured samples for training and evaluation under a grouped three-fold cross-subject evaluation framework. Five architectures were compared: standard MeshGraphNet, hierarchical MeshGraphNet, a routing-only transformer, a topology-biased routing transformer, and a hybrid model. The hybrid model achieved the best overall performance, yielding the lowest full-field error and peak stress error, together with the highest spatial agreement for high-risk regions. Among the non-hybrid models, the standard topology-diffusion model performed best overall, whereas routing-only strategies were less effective. These findings indicate that topology diffusion provides a robust basis for surrogate modeling of knee joint contact mechanics within the present benchmark, while the addition of global routing can further improve reconstruction of clinically relevant high-stress patterns." }, "2512.19528.pdf": { "id": "2512.19528v1", "title": "Multi-Modal Soccer Scene Analysis with Masked Pre-Training", "year": "2025", "authors": [ "Marc Peral", "Guillem Capellera", "Luis Ferraz", "Antonio Rubio", "Antonio Agudo" ], "categories": [ "cs.CV" ], "pdf_url": "https://arxiv.org/pdf/2512.19528v1", "abstract": "In this work we propose a multi-modal architecture for analyzing soccer scenes from tactical camera footage, with a focus on three core tasks: ball trajectory inference, ball state classification, and ball possessor identification. To this end, our solution integrates three distinct input modalities (player trajectories, player types and image crops of individual players) into a unified framework that processes spatial and temporal dynamics using a cascade of sociotemporal transformer blocks. Unlike prior methods, which rely heavily on accurate ball tracking or handcrafted heuristics, our approach infers the ball trajectory without direct access to its past or future positions, and robustly identifies the ball state and ball possessor under noisy or occluded conditions from real top league matches. We also introduce CropDrop, a modality-specific masking pre-training strategy that prevents over-reliance on image features and encourages the model to rely on cross-modal patterns during pre-training. We show the effectiveness of our approach on a large-scale dataset providing substantial improvements over state-of-the-art baselines in all tasks. Our results highlight the benefits of combining structured and visual cues in a transformer-based architecture, and the importance of realistic masking strategies in multi-modal learning." }, "1906.05029.pdf": { "id": "1906.05029v2", "title": "A Bayesian Approach to In-Game Win Probability in Soccer", "year": "2019", "authors": [ "Pieter Robberechts", "Jan Van Haaren", "Jesse Davis" ], "categories": [ "cs.LG", "stat.ML" ], "pdf_url": "https://arxiv.org/pdf/1906.05029v2", "abstract": "In-game win probability models, which provide a sports team's likelihood of winning at each point in a game based on historical observations, are becoming increasingly popular. In baseball, basketball and American football, they have become important tools to enhance fan experience, to evaluate in-game decision-making, and to inform coaching decisions. While equally relevant in soccer, the adoption of these models is held back by technical challenges arising from the low-scoring nature of the sport. In this paper, we introduce an in-game win probability model for soccer that addresses the shortcomings of existing models. First, we demonstrate that in-game win probability models for other sports struggle to provide accurate estimates for soccer, especially towards the end of a game. Second, we introduce a novel Bayesian statistical framework that estimates running win, tie and loss probabilities by leveraging a set of contextual game state features. An empirical evaluation on eight seasons of data for the top-five soccer leagues demonstrates that our framework provides well-calibrated probabilities. Furthermore, two use cases show its ability to enhance fan experience and to evaluate performance in crucial game situations." }, "2601.00421.pdf": { "id": "2601.00421v2", "title": "Can Semantic Methods Enhance Team Sports Tactics? A Methodology for Football with Broader Applications", "year": "2026", "authors": [ "Alessio Di Rubbo", "Mattia Neri", "Remo Pareschi", "Marco Pedroni", "Roberto Valtancoli", "Paolino Zica" ], "categories": [ "cs.AI" ], "pdf_url": "https://arxiv.org/pdf/2601.00421v2", "abstract": "This paper explores how semantic-space reasoning, traditionally used in computational linguistics, can be extended to tactical decision-making in team sports. Building on the analogy between texts and teams -- where players act as words and collective play conveys meaning -- the proposed methodology models tactical configurations as compositional semantic structures. Each player is represented as a multidimensional vector integrating technical, physical, and psychological attributes; team profiles are aggregated through contextual weighting into a higher-level semantic representation. Within this shared vector space, tactical templates such as high press, counterattack, or possession build-up are encoded analogously to linguistic concepts. Their alignment with team profiles is evaluated using vector-distance metrics, enabling the computation of tactical ``fit'' and opponent-exploitation potential. A Python-based prototype demonstrates how these methods can generate interpretable, dynamically adaptive strategy recommendations, accompanied by fine-grained diagnostic insights at the attribute level. Beyond football, the approach offers a generalizable framework for collective decision-making and performance optimization in team-based domains -- ranging from basketball and hockey to cooperative robotics and human-AI coordination systems. The paper concludes by outlining future directions toward real-world data integration, predictive simulation, and hybrid human-machine tactical intelligence." }, "2212.11041.pdf": { "id": "2212.11041v1", "title": "What should clubs monitor to predict future value of football players", "year": "2022", "authors": [ "Ali Baouan", "Elsa Bismuth", "Aur\u00e8le Bohbot", "S\u00e9bastien Coustou", "Mathieu Lacome", "Mathieu Rosenbaum" ], "categories": [ "stat.AP" ], "pdf_url": "https://arxiv.org/pdf/2212.11041v1", "abstract": "Huge amounts of money are invested every year by football clubs on transfers. For both growth and survival, it is crucial for recruiting departments to make smart choices when targeting players. Therefore, it is very important to identify the right parameters to monitor to predict market value. The following paper aims at determining the relevant features that successfully forecast future value for football players. Success is measured against their market value from TransferMarkt. To select prominent features, we use Lasso regressions and Random Forest algorithms. Some obvious variables are selected but we also observe some subtle dependencies between features and future market value. Finally, we rank the Golden Boy nominees using our forecasts and show our methodology can successfully compare football players based on their quality." }, "1710.02824.pdf": { "id": "1710.02824v2", "title": "Beating the bookies with their own numbers - and how the online sports betting market is rigged", "year": "2017", "authors": [ "Lisandro Kaunitz", "Shenjun Zhong", "Javier Kreiner" ], "categories": [ "stat.AP", "cs.OH", "stat.OT" ], "pdf_url": "https://arxiv.org/pdf/1710.02824v2", "abstract": "The online sports gambling industry employs teams of data analysts to build forecast models that turn the odds at sports games in their favour. While several betting strategies have been proposed to beat bookmakers, from expert prediction models and arbitrage strategies to odds bias exploitation, their returns have been inconsistent and it remains to be shown that a betting strategy can outperform the online sports betting market. We designed a strategy to beat football bookmakers with their own numbers. Instead of building a forecasting model to compete with bookmakers predictions, we exploited the probability information implicit in the odds publicly available in the marketplace to find bets with mispriced odds. Our strategy proved profitable in a 10-year historical simulation using closing odds, a 6-month historical simulation using minute to minute odds, and a 5-month period during which we staked real money with the bookmakers (we made code, data and models publicly available). Our results demonstrate that the football betting market is inefficient - bookmakers can be consistently beaten across thousands of games in both simulated environments and real-life betting. We provide a detailed description of our betting experience to illustrate how the sports gambling industry compensates these market inefficiencies with discriminatory practices against successful clients." }, "2609.38611.pdf": { "id": "2609.38611v2", "title": "Generalized Fiducial Inference for Hybrid Discrete-Continuous Count Data: Theory and Application", "year": "2026", "authors": [ "Kendall Thomas", "Jan Hannig" ], "categories": [ "stat.ME", "math.ST", "stat.AP" ], "pdf_url": "https://arxiv.org/pdf/2609.38611v2", "abstract": "High-frequency athlete-tracking data can be summarized as quantile cubes (Thomas and Hannig, 2026): counts of time spent in bins of velocity, acceleration, and movement angle. In our application, each athlete-match session has 100 such bins, with counts ranging from under a hundred to several thousand. Whether a bin's negative binomial distribution is well approximated by a normal distribution depends on its shape parameter, the product of its mean and dispersion, rather than on the count alone. We develop a hybrid generalized fiducial framework that uses exact negative binomial structural equations where the estimated shape is small and moment-matched normal approximations where it is large. This approach provides uncertainty quantification for hundreds of regression and dispersion parameters without specifying a prior distribution. Only the normal components contribute to the fiducial Jacobian, which has a closed-form block-diagonal structure, and we prove a Bernstein-von Mises theorem for the correctly specified hybrid model. Simulations show near-nominal coverage, agreement with a flat-prior Bayesian analysis, and the importance of partitioning by estimated shape rather than by the mean or observed responses. In 216 sessions from 17 professional women's soccer athletes, the estimated position effects concentrate at the velocity extremes, while the estimated age effects follow an acceleration gradient. However, an athlete-level permutation check indicates that these patterns are not distinguishable from those obtained under athlete-label permutations." }, "2604.04419.pdf": { "id": "2604.04419v1", "title": "BoxComm: Benchmarking Category-Aware Commentary Generation and Narration Rhythm in Boxing", "year": "2026", "authors": [ "Kaiwen Wang", "Kaili Zheng", "Rongrong Deng", "Yiming Shi", "Chenyi Guo", "Ji Wu" ], "categories": [ "cs.CV" ], "pdf_url": "https://arxiv.org/pdf/2604.04419v1", "abstract": "Recent multimodal large language models (MLLMs) have shown strong capabilities in general video understanding, driving growing interest in automatic sports commentary generation. However, existing benchmarks for this task focus exclusively on team sports such as soccer and basketball, leaving combat sports entirely unexplored. Notably, combat sports present distinct challenges: critical actions unfold within milliseconds with visually subtle yet semantically decisive differences, and professional commentary contains a substantially higher proportion of tactical analysis compared to team sports. In this paper, we present BoxComm, a large-scale dataset comprising 445 World Boxing Championship match videos with over 52K commentary sentences from professional broadcasts. We propose a structured commentary taxonomy that categorizes each sentence into play-by-play, tactical, or contextual, providing the first category-level annotation for sports commentary benchmarks. Building on this taxonomy, we introduce two novel and complementary evaluations tailored to sports commentary generation: (1) category-conditioned generation, which evaluates whether models can produce accurate commentary of a specified type given video context; and (2) commentary rhythm assessment, which measures whether freely generated commentary exhibits appropriate temporal pacing and type distribution over continuous video segments, capturing a dimension of commentary competence that prior benchmarks have not addressed. Experiments on multiple state-of-the-art MLLMs reveal that current models struggle on both evaluations. We further propose EIC-Gen, an improved baseline incorporating detected punch events to supply structured action cues, yielding consistent gains and highlighting the importance of perceiving fleeting and subtle events for combat sports commentary." }, "2509.20083.pdf": { "id": "2509.20083v2", "title": "Rethinking player evaluation in sports: Goals above expectation and beyond", "year": "2025", "authors": [ "Robert Bajons", "Lucas Kook" ], "categories": [ "stat.AP" ], "pdf_url": "https://arxiv.org/pdf/2509.20083v2", "abstract": "A popular quantitative approach to evaluating player performance in sports involves comparing an observed outcome to the expected outcome ignoring player involvement, which is estimated using statistical or machine learning methods. In soccer, for instance, goals above expectation (GAX) of a player measure how often shots of this player led to a goal compared to the model-derived expected outcome of the shots. Typically, sports data analysts rely on flexible machine learning models, which are capable of handling complex nonlinear effects and feature interactions, but fail to provide valid statistical inference due to finite-sample bias and slow convergence rates. In this paper, we close this gap by presenting a framework for player evaluation with metrics derived from differences in actual and expected outcomes using flexible machine learning algorithms, which nonetheless allows for valid frequentist inference. We first show that the commonly used metrics are directly related to Rao's score test in parametric regression models for the expected outcome. Motivated by this finding and recent developments in double machine learning, we then propose the use of residualized versions of the original metrics. For GAX, the residualization step corresponds to an additional regression predicting whether a given player would take the shot under the circumstances described by the features. We further relate metrics in the proposed framework to player-specific effect estimates in interpretable semiparametric regression models, allowing us to infer directional effects, e.g., to determine players that have a positive impact on the outcome. Our primary use case are GAX in soccer. We further apply our framework to evaluate goal-stopping ability of goalkeepers, shooting skill in basketball, quarterback passing skill in American football, and injury-proneness of soccer players." }, "2507.22782.pdf": { "id": "2507.22782v3", "title": "Enhancing Multi-Agent Collaboration with Attention-Based Actor-Critic Policies", "year": "2025", "authors": [ "Hugo Garrido-Lestache Belinchon", "Jeremy Kedziora" ], "categories": [ "cs.AI", "cs.LG" ], "pdf_url": "https://arxiv.org/pdf/2507.22782v3", "abstract": "This paper introduces Team-Attention-Actor-Critic (TAAC), a reinforcement learning algorithm designed to enhance multi-agent collaboration in cooperative environments. TAAC employs a Centralized Training/Centralized Execution scheme incorporating multi-headed attention mechanisms in both the actor and critic. This design facilitates dynamic, inter-agent communication, allowing agents to explicitly query teammates, thereby efficiently managing the exponential growth of joint-action spaces while ensuring a high degree of collaboration. We further introduce a penalized loss function which promotes diverse yet complementary roles among agents. We evaluate TAAC in a simulated soccer environment against benchmark algorithms representing other multi-agent paradigms, including Proximal Policy Optimization and Multi-Agent Actor-Attention-Critic. We find that TAAC exhibits superior performance and enhanced collaborative behaviors across a variety of metrics (win rates, goal differentials, Elo ratings, inter-agent connectivity, balanced spatial distributions, and frequent tactical interactions such as ball possession swaps)." }, "2310.11459.pdf": { "id": "2310.11459v1", "title": "Ratings of European and South American Football Leagues Based on Glicko-2 with Modifications", "year": "2023", "authors": [ "Andrei Shelopugin", "Alexander Sirotkin" ], "categories": [ "stat.AP" ], "pdf_url": "https://arxiv.org/pdf/2310.11459v1", "abstract": "One of the key problems in the field of soccer analytics is predicting how a player performance changes when transitioning from one league to another. One potential solution to address this issue lies in the evaluation of the respective league strength. This article endeavors to compute club ratings of the first and second European and South American leagues. In order to calculate these ratings, the authors have designed the Glicko-2 rating system-based approach, which overcomes some Glicko-2 limitations. Particularly, the authors took into consideration the probability of the draw, the home-field advantage, and the property of teams to become stronger or weaker following their league transitions. Furthermore, authors have constructed a predictive model for forecasting match results based on the number of goals scored in previous matches. The metrics utilized in the analysis reveal that the Glicko-2 based approach exhibits a marginally superior level of accuracy when compared to the commonly used Poisson regression-based approach. In addition, Glicko-2 based ratings offer greater interpretability and can find application in various analytics tasks, such as predicting soccer player metrics for forthcoming seasons or the detailed analysis of a player performance in preceding matches. The implementation of the approach is available on github.com/andreyshelopugin/GlickoSoccer." }, "2307.13700.pdf": { "id": "2307.13700v1", "title": "CAMP: A Context-Aware Cricket Players Performance Metric", "year": "2023", "authors": [ "Muhammad Sohaib Ayub", "Naimat Ullah", "Sarwan Ali", "Imdad Ullah Khan", "Mian Muhammad Awais", "Muhammad Asad Khan", "Safiullah Faizullah" ], "categories": [ "cs.AI", "cs.CY", "cs.LG" ], "pdf_url": "https://arxiv.org/pdf/2307.13700v1", "abstract": "Cricket is the second most popular sport after soccer in terms of viewership. However, the assessment of individual player performance, a fundamental task in team sports, is currently primarily based on aggregate performance statistics, including average runs and wickets taken. We propose Context-Aware Metric of player Performance, CAMP, to quantify individual players' contributions toward a cricket match outcome. CAMP employs data mining methods and enables effective data-driven decision-making for selection and drafting, coaching and training, team line-ups, and strategy development. CAMP incorporates the exact context of performance, such as opponents' strengths and specific circumstances of games, such as pressure situations. We empirically evaluate CAMP on data of limited-over cricket matches between 2001 and 2019. In every match, a committee of experts declares one player as the best player, called Man of the M}atch (MoM). The top two rated players by CAMP match with MoM in 83\\% of the 961 games. Thus, the CAMP rating of the best player closely matches that of the domain experts. By this measure, CAMP significantly outperforms the current best-known players' contribution measure based on the Duckworth-Lewis-Stern (DLS) method." }, "2005.12553.pdf": { "id": "2005.12553v1", "title": "Efficient Use of heuristics for accelerating XCS-based Policy Learning in Markov Games", "year": "2020", "authors": [ "Hao Chen", "Chang Wang", "Jian Huang", "Jianxing Gong" ], "categories": [ "cs.AI", "cs.GT", "cs.MA", "cs.NE" ], "pdf_url": "https://arxiv.org/pdf/2005.12553v1", "abstract": "In Markov games, playing against non-stationary opponents with learning ability is still challenging for reinforcement learning (RL) agents, because the opponents can evolve their policies concurrently. This increases the complexity of the learning task and slows down the learning speed of the RL agents. This paper proposes efficient use of rough heuristics to speed up policy learning when playing against concurrent learners. Specifically, we propose an algorithm that can efficiently learn explainable and generalized action selection rules by taking advantages of the representation of quantitative heuristics and an opponent model with an eXtended classifier system (XCS) in zero-sum Markov games. A neural network is used to model the opponent from their behaviors and the corresponding policy is inferred for action selection and rule evolution. In cases of multiple heuristic policies, we introduce the concept of Pareto optimality for action selection. Besides, taking advantages of the condition representation and matching mechanism of XCS, the heuristic policies and the opponent model can provide guidance for situations with similar feature representation. Furthermore, we introduce an accuracy-based eligibility trace mechanism to speed up rule evolution, i.e., classifiers that can match the historical traces are reinforced according to their accuracy. We demonstrate the advantages of the proposed algorithm over several benchmark algorithms in a soccer and a thief-and-hunter scenarios." }, "1909.09849.pdf": { "id": "1909.09849v4", "title": "Multiagent Evaluation under Incomplete Information", "year": "2019", "authors": [ "Mark Rowland", "Shayegan Omidshafiei", "Karl Tuyls", "Julien Perolat", "Michal Valko", "Georgios Piliouras", "Remi Munos" ], "categories": [ "cs.MA", "cs.AI", "cs.LG", "stat.ML" ], "pdf_url": "https://arxiv.org/pdf/1909.09849v4", "abstract": "This paper investigates the evaluation of learned multiagent strategies in the incomplete information setting, which plays a critical role in ranking and training of agents. Traditionally, researchers have relied on Elo ratings for this purpose, with recent works also using methods based on Nash equilibria. Unfortunately, Elo is unable to handle intransitive agent interactions, and other techniques are restricted to zero-sum, two-player settings or are limited by the fact that the Nash equilibrium is intractable to compute. Recently, a ranking method called \u03b1-Rank, relying on a new graph-based game-theoretic solution concept, was shown to tractably apply to general games. However, evaluations based on Elo or \u03b1-Rank typically assume noise-free game outcomes, despite the data often being collected from noisy simulations, making this assumption unrealistic in practice. This paper investigates multiagent evaluation in the incomplete information regime, involving general-sum many-player games with noisy outcomes. We derive sample complexity guarantees required to confidently rank agents in this setting. We propose adaptive algorithms for accurate ranking, provide correctness and sample complexity guarantees, then introduce a means of connecting uncertainties in noisy match outcomes to uncertainties in rankings. We evaluate the performance of these approaches in several domains, including Bernoulli games, a soccer meta-game, and Kuhn poker." }, "1906.01131.pdf": { "id": "1906.01131v1", "title": "Hybrid Machine Learning Forecasts for the FIFA Women's World Cup 2019", "year": "2019", "authors": [ "Andreas Groll", "Christophe Ley", "Gunther Schauberger", "Hans Van Eetvelde", "Achim Zeileis" ], "categories": [ "stat.ML", "cs.LG", "stat.AP" ], "pdf_url": "https://arxiv.org/pdf/1906.01131v1", "abstract": "In this work, we combine two different ranking methods together with several other predictors in a joint random forest approach for the scores of soccer matches. The first ranking method is based on the bookmaker consensus, the second ranking method estimates adequate ability parameters that reflect the current strength of the teams best. The proposed combined approach is then applied to the data from the two previous FIFA Women's World Cups 2011 and 2015. Finally, based on the resulting estimates, the FIFA Women's World Cup 2019 is simulated repeatedly and winning probabilities are obtained for all teams. The model clearly favors the defending champion USA before the host France." }, "2608.04008.pdf": { "id": "2608.04008v1", "title": "WorldCup Arena: Prospective, Leakage-Free Evaluation of Frontier LLMs on a Live Tournament", "year": "2026", "authors": [ "Zhenran Wang", "Zhonghan Bian", "Jinsong Li", "Zhangyang Qi" ], "categories": [ "cs.CL" ], "pdf_url": "https://arxiv.org/pdf/2608.04008v1", "abstract": "Benchmarks that measure the forecasting ability of large language models are almost always retrospective: the event has happened, the answer is somewhere on the Web, and the evaluation must defend itself against memorisation. We report the opposite design. Over the 39 days of the 2026 FIFA World Cup, six frontier LLMs -- all with extended thinking and native server-side web search -- were asked before every kickoff, one match at a time, to fill in a seven-market prediction card for all 104 matches, plus 12 group winners and a pre-tournament outright pool; no answer existed when the question was asked, so the evaluation is leakage-free by construction rather than by filtering, and the frozen archive holds 4,494 scored predictions. What the tournament establishes is a set of behaviours the six systems share. On match outcome they average 63.9%, level with backing the bookmaker's favourite -- which is in fact what they usually do. They agree with one another far more often than they are right, so a majority vote adds nothing. They under-commit to draws and to goals, and crowd their scoreline picks onto a single prototypical result. Accuracy tracks how lopsided a fixture is rather than how much is known about it: it collapses in the closest ties, where the dossiers are richest, while questions about the tournament as a whole are answered well. On this task the current generation of frontier systems is not sharply differentiated: the standings hold up at the top and the bottom across the run and churn in the middle, and the margins stay narrow throughout. The briefing dossiers, fixtures and official results are released as a benchmark, together with the scoring code." }, "2512.18013.pdf": { "id": "2512.18013v1", "title": "Empirical parameterization of the Elo Rating System", "year": "2025", "authors": [ "Shirsa Maitra", "Tathagata Banerjee", "Anushka De", "Diganta Mukherjee", "Tridib Mukherjee" ], "categories": [ "stat.AP" ], "pdf_url": "https://arxiv.org/pdf/2512.18013v1", "abstract": "This study aims to provide a data-driven approach for empirically tuning and validating rating systems, focusing on the Elo system. Well-known rating frameworks, such as Elo, Glicko, TrueSkill systems, rely on parameters that are usually chosen based on probabilistic assumptions or conventions, and do not utilize game-specific data. To address this issue, we propose a methodology that learns optimal parameter values by maximizing the predictive accuracy of match outcomes. The proposed parameter-tuning framework is a generalizable method that can be extended to any rating system, even for multiplayer setups, through suitable modification of the parameter space. Implementation of the rating system on real and simulated gameplay data demonstrates the suitability of the data-driven rating system in modeling player performance." }, "1802.08848.pdf": { "id": "1802.08848v1", "title": "Combining historical data and bookmakers'odds in modelling football scores", "year": "2018", "authors": [ "Leonardo Egidi", "Francesco Pauli", "Nicola Torelli" ], "categories": [ "stat.AP" ], "pdf_url": "https://arxiv.org/pdf/1802.08848v1", "abstract": "Modelling football outcomes has gained increasing attention, in large part due to the potential for making substantial profits. Despite the strong connection existing between football models and the bookmakers' betting odds, no authors have used the latter for improving the fit and the predictive accuracy of these models. We have developed a hierarchical Bayesian Poisson model in which the scoring rates of the teams are convex combinations of parameters estimated from historical data and the additional source of the betting odds. We apply our analysis to a nine-year dataset of the most popular European leagues in order to predict match outcomes for their tenth seasons. In this paper, we provide numerical and graphical checks for our model." }, "2605.13926.pdf": { "id": "2605.13926v1", "title": "Optimising football transfer strategy under budget constraints: A weighted multi-criteria approach", "year": "2026", "authors": [ "Tathagata Basu", "Soudeep Deb", "Rishideep Roy" ], "categories": [ "stat.AP" ], "pdf_url": "https://arxiv.org/pdf/2605.13926v1", "abstract": "The football transfer market is a complex, dynamic environment in which clubs compete to acquire players who strengthen their squads. While several frameworks estimate a player's worth, a comprehensive approach that captures both squad optimisation and transfer market dynamics remains limited. In this paper, we propose a quantitative framework for optimising football transfer strategy under budget constraints, integrated with a competitive bidding paradigm. Using data from professional football leagues, we construct player performance and transfer price models using linear mixed-effects frameworks that incorporate player characteristics, recent performance, team context, and league effects. The predicted ratings and estimated transfer prices are then integrated into a weighted multi-criteria constrained optimisation framework that determines a club's transfer activities at the end of the season. Finally, these optimal transfer decisions are embedded within an independent private-value auction model with a random reserve price to analyse market behaviour when multiple teams compete for the same player. We illustrate our approach using the 2018-19 season of the English Premier League to demonstrate its ability to capture transfer-market dynamics." }, "2606.11120.pdf": { "id": "2606.11120v1", "title": "Monte Carlo Pass Search: Using Trajectory Generation for 3D Counterfactual Pass Evaluation in Football", "year": "2026", "authors": [ "Andrew Kang", "Priya Narasimhan" ], "categories": [ "cs.AI", "cs.CV" ], "pdf_url": "https://arxiv.org/pdf/2606.11120v1", "abstract": "We recast pass evaluation in football (soccer) as a Monte Carlo Tree Search (MCTS)-like evaluation problem whose components mostly exist in the literature under different names: a value model (possession value), a world model (multi-agent trajectories with ball interactions), and a policy over counterfactual actions (sampling pass variants with noise). Building on the first public high-fidelity tracking dataset with 3D ball trajectories from the Bundesliga, we introduce Monte Carlo Pass Search (MCPS), which infers kick parameters for each observed pass, samples execution variants and option variants, rolls each candidate forward with a ball-conditioned world model until the next ball interaction, and scores outcomes with a learned value model to obtain a distribution over gained value. This distribution enables distribution-aware attribution with two complementary execution-surplus scores used for analysis and ranking: mean-based and percentile-based scores. To make the world model sample-efficient under limited public data, we adapt a discrete-token, autoregressive trajectory generator from autonomous driving (SMART) and show it yields strong best-of-20 forecasting accuracy compared to baselines, while supporting fully hypothetical rollouts for downstream evaluation. We have released model checkpoints and code." }, "2604.19104.pdf": { "id": "2604.19104v1", "title": "Reinforcement Learning Enabled Adaptive Multi-Task Control for Bipedal Soccer Robots", "year": "2026", "authors": [ "Yulai Zhang", "Yinrong Zhang", "Ting Wu", "Linqi Ye" ], "categories": [ "cs.RO", "cs.AI" ], "pdf_url": "https://arxiv.org/pdf/2604.19104v1", "abstract": "Developing bipedal football robots in dynamiccombat environments presents challenges related to motionstability and deep coupling of multiple tasks, as well ascontrol switching issues between different states such as up-right walking and fall recovery. To address these problems,this paper proposes a modular reinforcement learning (RL)framework for achieving adaptive multi-task control. Firstly,this framework combines an open-loop feedforward oscilla-tor with a reinforcement learning-based feedback residualstrategy, effectively separating the generation of basic gaitsfrom complex football actions. Secondly, a posture-driven statemachine is introduced, clearly switching between the ballseeking and kicking network (BSKN) and the fall recoverynetwork (FRN), fundamentally preventing state interference.The FRN is efficiently trained through a progressive forceattenuation curriculum learning strategy. The architecture wasverified in Unity simulations of bipedal robots, demonstratingexcellent spatial adaptability-reliably finding and kicking theball even in restricted corner scenarios-and rapid autonomousfall recovery (with an average recovery time of 0.715 seconds).This ensures seamless and stable operation in complex multi-task environments." }, "2512.10355.pdf": { "id": "2512.10355v1", "title": "Better Prevent than Tackle: Valuing Defense in Soccer Based on Graph Neural Networks", "year": "2025", "authors": [ "Hyunsung Kim", "Sangwoo Seo", "Hoyoung Choi", "Tom Boomstra", "Jinsung Yoon", "Chanyoung Park" ], "categories": [ "cs.LG", "cs.MA" ], "pdf_url": "https://arxiv.org/pdf/2512.10355v1", "abstract": "Evaluating defensive performance in soccer remains challenging, as effective defending is often expressed not through visible on-ball actions such as interceptions and tackles, but through preventing dangerous opportunities before they arise. Existing approaches have largely focused on valuing on-ball actions, leaving much of defenders' true impact unmeasured. To address this gap, we propose DEFCON (DEFensive CONtribution evaluator), a comprehensive framework that quantifies player-level defensive contributions for every attacking situation in soccer. Leveraging Graph Attention Networks, DEFCON estimates the success probability and expected value of each attacking option, along with each defender's responsibility for stopping it. These components yield an Expected Possession Value (EPV) for the attacking team before and after each action, and DEFCON assigns positive or negative credits to defenders according to whether they reduced or increased the opponent's EPV. Trained on 2023-24 and evaluated on 2024-25 Eredivisie event and tracking data, DEFCON's aggregated player credits exhibit strong positive correlations with market valuations. Finally, we showcase several practical applications, including in-game timelines of defensive contributions, spatial analyses across pitch zones, and pairwise summaries of attacker-defender interactions." }, "2512.03166.pdf": { "id": "2512.03166v1", "title": "Multi-Agent Reinforcement Learning and Real-Time Decision-Making in Robotic Soccer for Virtual Environments", "year": "2025", "authors": [ "Aya Taourirte", "Md Sohag Mia" ], "categories": [ "cs.RO", "cs.CV" ], "pdf_url": "https://arxiv.org/pdf/2512.03166v1", "abstract": "The deployment of multi-agent systems in dynamic, adversarial environments like robotic soccer necessitates real-time decision-making, sophisticated cooperation, and scalable algorithms to avoid the curse of dimensionality. While Reinforcement Learning (RL) offers a promising framework, existing methods often struggle with the multi-granularity of tasks (long-term strategy vs. instant actions) and the complexity of large-scale agent interactions. This paper presents a unified Multi-Agent Reinforcement Learning (MARL) framework that addresses these challenges. First, we establish a baseline using Proximal Policy Optimization (PPO) within a client-server architecture for real-time action scheduling, with PPO demonstrating superior performance (4.32 avg. goals, 82.9% ball control). Second, we introduce a Hierarchical RL (HRL) structure based on the options framework to decompose the problem into a high-level trajectory planning layer (modeled as a Semi-Markov Decision Process) and a low-level action execution layer, improving global strategy (avg. goals increased to 5.26). Finally, to ensure scalability, we integrate mean-field theory into the HRL framework, simplifying many-agent interactions into a single agent vs. the population average. Our mean-field actor-critic method achieves a significant performance boost (5.93 avg. goals, 89.1% ball control, 92.3% passing accuracy) and enhanced training stability. Extensive simulations of 4v4 matches in the Webots environment validate our approach, demonstrating its potential for robust, scalable, and cooperative behavior in complex multi-agent domains." }, "2511.00261.pdf": { "id": "2511.00261v2", "title": "Spot The Ball: A Benchmark for Visual Social Inference", "year": "2025", "authors": [ "Neha Balamurugan", "Sarah Wu", "Adam Chun", "Gabe Gaw", "Cristobal Eyzaguirre", "Tobias Gerstenberg" ], "categories": [ "cs.CV", "cs.HC" ], "pdf_url": "https://arxiv.org/pdf/2511.00261v2", "abstract": "Humans excel at visual social inference, the ability to infer hidden elements of a scene from subtle behavioral cues such as other people's gaze, pose, and orientation. This ability drives everyday social reasoning in humans and is critical for developing more human-like AI agents. We introduce Spot The Ball, a challenging benchmark for evaluating visual social inference in vision-language models (VLMs) using sports as a test domain. The task is to localize a removed sports ball from soccer, basketball, and volleyball images. We present a curated evaluation set with human baselines and a scalable pipeline for generating additional test items. We evaluate four state-of-the-art VLMs (Gemini, GPT, LLaMA, Qwen) using three prompting strategies, finding that humans are consistently two to three times more accurate (20-34%) than models ($\\leq$ 17%) across all sports. Our analyses show that models rely on superficial spatial heuristics--such as guessing near the image center or nearby players--while humans leverage social cues like gaze direction and body pose. These findings reveal a persistent human-model gap in visual social reasoning and underscore the need for architectures that explicitly encode structured behavioral cues to achieve robust, human-like inference." }, "2509.14477.pdf": { "id": "2509.14477v1", "title": "Ticket-Bench: A Kickoff for Multilingual and Regionalized Agent Evaluation", "year": "2025", "authors": [ "Thales Sales Almeida", "Jo\u00e3o Guilherme Alves Santos", "Thiago Laitz", "Giovana Kerche Bon\u00e1s" ], "categories": [ "cs.CL" ], "pdf_url": "https://arxiv.org/pdf/2509.14477v1", "abstract": "Large language models (LLMs) are increasingly deployed as task-oriented agents, where success depends on their ability to generate accurate function calls under realistic, multilingual conditions. However, existing agent evaluations largely overlook cultural and linguistic diversity, often relying on monolingual or naively translated benchmarks. We introduce Ticket-Bench, a benchmark for multilingual agent evaluation in task-oriented scenarios. Ticket-Bench simulates the domain of soccer ticket purchases across six major languages: Portuguese, English, Spanish, German, Italian, and French. Using localized teams, cities, and user profiles to provide a higher level of realism. We evaluate a wide range of commercial and open-source LLMs, measuring function-calling accuracy and consistency across languages. Results show that reasoning-oriented models (e.g., GPT-5, Qwen3-235B) dominate performance but still exhibit notable cross-lingual disparities. These findings underscore the need for culturally aware, multilingual benchmarks to guide the development of robust LLM agents." }, "2505.16199.pdf": { "id": "2505.16199v1", "title": "Velocity Completion Task and Method for Event-based Player Positional Data in Soccer", "year": "2025", "authors": [ "Rikuhei Umemoto", "Keisuke Fujii" ], "categories": [ "cs.AI" ], "pdf_url": "https://arxiv.org/pdf/2505.16199v1", "abstract": "In many real-world complex systems, the behavior can be observed as a collection of discrete events generated by multiple interacting agents. Analyzing the dynamics of these multi-agent systems, especially team sports, often relies on understanding the movement and interactions of individual agents. However, while providing valuable snapshots, event-based positional data typically lacks the continuous temporal information needed to directly calculate crucial properties such as velocity. This absence severely limits the depth of dynamic analysis, preventing a comprehensive understanding of individual agent behaviors and emergent team strategies. To address this challenge, we propose a new method to simultaneously complete the velocity of all agents using only the event-based positional data from team sports. Based on this completed velocity information, we investigate the applicability of existing team sports analysis and evaluation methods. Experiments using soccer event data demonstrate that neural network-based approaches outperformed rule-based methods regarding velocity completion error, considering the underlying temporal dependencies and graph structure of player-to-player or player-to-ball interaction. Moreover, the space evaluation results obtained using the completed velocity are closer to those derived from complete tracking data, highlighting our method's potential for enhanced team sports system analysis." }, "2404.09951.pdf": { "id": "2404.09951v1", "title": "Unifying Global and Local Scene Entities Modelling for Precise Action Spotting", "year": "2024", "authors": [ "Kim Hoang Tran", "Phuc Vuong Do", "Ngoc Quoc Ly", "Ngan Le" ], "categories": [ "cs.CV" ], "pdf_url": "https://arxiv.org/pdf/2404.09951v1", "abstract": "Sports videos pose complex challenges, including cluttered backgrounds, camera angle changes, small action-representing objects, and imbalanced action class distribution. Existing methods for detecting actions in sports videos heavily rely on global features, utilizing a backbone network as a black box that encompasses the entire spatial frame. However, these approaches tend to overlook the nuances of the scene and struggle with detecting actions that occupy a small portion of the frame. In particular, they face difficulties when dealing with action classes involving small objects, such as balls or yellow/red cards in soccer, which only occupy a fraction of the screen space. To address these challenges, we introduce a novel approach that analyzes and models scene entities using an adaptive attention mechanism. Particularly, our model disentangles the scene content into the global environment feature and local relevant scene entities feature. To efficiently extract environmental features while considering temporal information with less computational cost, we propose the use of a 2D backbone network with a time-shift mechanism. To accurately capture relevant scene entities, we employ a Vision-Language model in conjunction with the adaptive attention mechanism. Our model has demonstrated outstanding performance, securing the 1st place in the SoccerNet-v2 Action Spotting, FineDiving, and FineGym challenge with a substantial performance improvement of 1.6, 2.0, and 1.3 points in avg-mAP compared to the runner-up methods. Furthermore, our approach offers interpretability capabilities in contrast to other deep learning models, which are often designed as black boxes. Our code and models are released at: https://github.com/Fsoft-AIC/unifying-global-local-feature." }, "2401.16235.pdf": { "id": "2401.16235v2", "title": "Player Pressure Map -- A Novel Representation of Pressure in Soccer for Evaluating Player Performance in Different Game Contexts", "year": "2024", "authors": [ "Chaoyi Gu", "Jiaming Na", "Yisheng Pei", "Varuna De Silva" ], "categories": [ "cs.LG", "stat.AP" ], "pdf_url": "https://arxiv.org/pdf/2401.16235v2", "abstract": "In soccer, contextual player performance metrics are invaluable to coaches. For example, the ability to perform under pressure during matches distinguishes the elite from the average. Appropriate pressure metric enables teams to assess players' performance accurately under pressure and design targeted training scenarios to address their weaknesses. The primary objective of this paper is to leverage both tracking and event data and game footage to capture the pressure experienced by the possession team in a soccer game scene. We propose a player pressure map to represent a given game scene, which lowers the dimension of raw data and still contains rich contextual information. Not only does it serve as an effective tool for visualizing and evaluating the pressure on the team and each individual, but it can also be utilized as a backbone for accessing players' performance. Overall, our model provides coaches and analysts with a deeper understanding of players' performance under pressure so that they make data-oriented tactical decisions." }, "2312.14360.pdf": { "id": "2312.14360v2", "title": "Designing a skilled soccer team for RoboCup: exploring skill-set-primitives through reinforcement learning", "year": "2023", "authors": [ "Miguel Abreu", "Luis Paulo Reis", "Nuno Lau" ], "categories": [ "cs.RO" ], "pdf_url": "https://arxiv.org/pdf/2312.14360v2", "abstract": "The RoboCup 3D Soccer Simulation League serves as a competitive platform for showcasing innovation in autonomous humanoid robot agents through simulated soccer matches. Our team, FC Portugal, developed a new codebase from scratch in Python after RoboCup 2021. The team's performance relies on a set of skills centered around novel unifying primitives and a custom, symmetry-extended version of the Proximal Policy Optimization algorithm. Our methods have been thoroughly tested in official RoboCup matches, where FC Portugal has won the last two main competitions, in 2022 and 2023. This paper presents our training framework, as well as a timeline of skills developed using our skill-set-primitives, which considerably improve the sample efficiency and stability of skills, and motivate seamless transitions. We start with a significantly fast sprint-kick developed in 2021 and progress to the most recent skill set, including a multi-purpose omnidirectional walk, a dribble with unprecedented ball control, a solid kick, and a push skill. The push addresses low-level collision scenarios and high-level strategies to increase ball possession. We address the resource-intensive nature of this task through an innovative multi-agent learning approach. Finally, we release the team's codebase to the RoboCup community, providing other teams with a robust and modern foundation upon which they can build new features." }, "2202.05728.pdf": { "id": "2202.05728v2", "title": "Deep soccer captioning with transformer: dataset, semantics-related losses, and multi-level evaluation", "year": "2022", "authors": [ "Ahmad Hammoudeh", "Bastien Vanderplaetse", "St\u00e9phane Dupont" ], "categories": [ "cs.CV", "cs.AI" ], "pdf_url": "https://arxiv.org/pdf/2202.05728v2", "abstract": "This work aims at generating captions for soccer videos using deep learning. In this context, this paper introduces a dataset, model, and triple-level evaluation. The dataset consists of 22k caption-clip pairs and three visual features (images, optical flow, inpainting) for ~500 hours of \\emph{SoccerNet} videos. The model is divided into three parts: a transformer learns language, ConvNets learn vision, and a fusion of linguistic and visual features generates captions. The paper suggests evaluating generated captions at three levels: syntax (the commonly used evaluation metrics such as BLEU-score and CIDEr), meaning (the quality of descriptions for a domain expert), and corpus (the diversity of generated captions). The paper shows that the diversity of generated captions has improved (from 0.07 reaching 0.18) with semantics-related losses that prioritize selected words. Semantics-related losses and the utilization of more visual features (optical flow, inpainting) improved the normalized captioning score by 28\\%. The web page of this work: https://sites.google.com/view/soccercaptioning}{https://sites.google.com/view/soccercaptioning" }, "2102.09469.pdf": { "id": "2102.09469v1", "title": "Optimising Long-Term Outcomes using Real-World Fluent Objectives: An Application to Football", "year": "2021", "authors": [ "Ryan Beal", "Georgios Chalkiadakis", "Timothy J. Norman", "Sarvapali D. Ramchurn" ], "categories": [ "cs.AI", "cs.GT" ], "pdf_url": "https://arxiv.org/pdf/2102.09469v1", "abstract": "In this paper, we present a novel approach for optimising long-term tactical and strategic decision-making in football (soccer) by encapsulating events in a league environment across a given time frame. We model the teams' objectives for a season and track how these evolve as games unfold to give a fluent objective that can aid in decision-making games. We develop Markov chain Monte Carlo and deep learning-based algorithms that make use of the fluent objectives in order to learn from prior games and other games in the environment and increase the teams' long-term performance. Simulations of our approach using real-world datasets from 760 matches shows that by using optimised tactics with our fluent objective and prior games, we can on average increase teams mean expected finishing distribution in the league by up to 35.6%." }, "1911.01529.pdf": { "id": "1911.01529v2", "title": "Closing the Reality Gap with Unsupervised Sim-to-Real Image Translation", "year": "2019", "authors": [ "Jan Blumenkamp", "Andreas Baude", "Tim Laue" ], "categories": [ "cs.LG", "cs.CV", "cs.RO" ], "pdf_url": "https://arxiv.org/pdf/1911.01529v2", "abstract": "Deep learning approaches have become the standard solution to many problems in computer vision and robotics, but obtaining sufficient training data in high enough quality is challenging, as human labor is error prone, time consuming, and expensive. Solutions based on simulation have become more popular in recent years, but the gap between simulation and reality is still a major issue. In this paper, we introduce a novel method for augmenting synthetic image data through unsupervised image-to-image translation by applying the style of real world images to simulated images with open source frameworks. The generated dataset is combined with conventional augmentation methods and is then applied to a neural network model running in real-time on autonomous soccer robots. Our evaluation shows a significant improvement compared to models trained on images generated entirely in simulation." }, "1809.04560.pdf": { "id": "1809.04560v2", "title": "Game-Based Video-Context Dialogue", "year": "2018", "authors": [ "Ramakanth Pasunuru", "Mohit Bansal" ], "categories": [ "cs.CL", "cs.AI", "cs.CV" ], "pdf_url": "https://arxiv.org/pdf/1809.04560v2", "abstract": "Current dialogue systems focus more on textual and speech context knowledge and are usually based on two speakers. Some recent work has investigated static image-based dialogue. However, several real-world human interactions also involve dynamic visual context (similar to videos) as well as dialogue exchanges among multiple speakers. To move closer towards such multimodal conversational skills and visually-situated applications, we introduce a new video-context, many-speaker dialogue dataset based on live-broadcast soccer game videos and chats from Twitch.tv. This challenging testbed allows us to develop visually-grounded dialogue models that should generate relevant temporal and spatial event language from the live video, while also being relevant to the chat history. For strong baselines, we also present several discriminative and generative models, e.g., based on tridirectional attention flow (TriDAF). We evaluate these models via retrieval ranking-recall, automatic phrase-matching metrics, as well as human evaluation studies. We also present dataset analyses, model ablations, and visualizations to understand the contribution of different modalities and model components." }, "1612.00947.pdf": { "id": "1612.00947v1", "title": "Disruptive innovations in RoboCup 2D Soccer Simulation League: from Cyberoos'98 to Gliders2016", "year": "2016", "authors": [ "Mikhail Prokopenko", "Peter Wang" ], "categories": [ "cs.MA" ], "pdf_url": "https://arxiv.org/pdf/1612.00947v1", "abstract": "We review disruptive innovations introduced in the RoboCup 2D Soccer Simulation League over the twenty years since its inception, and trace the progress of our champion team (Gliders). We conjecture that the League has been developing as an ecosystem shaped by diverse approaches taken by participating teams, increasing in its overall complexity. A common feature is that different champion teams succeeded in finding a way to decompose the enormous search-space of possible single- and multi-agent behaviours, by automating the exploration of the problem space with various techniques which accelerated the software development efforts. These methods included interactive debugging, machine learning, automated planning, and opponent modelling. The winning approach developed by Gliders is centred on human-based evolutionary computation which optimised several components such as an action-dependent evaluation function, dynamic tactics with Voronoi diagrams, information dynamics, and bio-inspired collective behaviour." }, "2001.09097.pdf": { "id": "2001.09097v1", "title": "Forecasting football matches by predicting match statistics", "year": "2020", "authors": [ "Edward Wheatcroft" ], "categories": [ "stat.AP" ], "pdf_url": "https://arxiv.org/pdf/2001.09097v1", "abstract": "This paper considers the use of observed and predicted match statistics as inputs to forecasts of the outcomes of football matches. It is shown that, were it possible to know the match statistics in advance, highly informative forecasts of the match outcome could be made. Whilst, in practice, match statistics are never available prior to the match, this leads to a simple philosophy. If match statistics can be predicted pre-match, and if those predictions are accurate enough, it follows that informative match forecasts can be made. It is shown that Generalised Attacking Performance (GAP) ratings, introduced in a recent paper, provide a suitable methodology for predicting match statistics in advance and that they are informative enough to provide information beyond that reflected in the odds. A long term and robust gambling profit is demonstrated when the forecasts are combined with two betting strategies." }, "2610.22444.pdf": { "id": "2610.22444v1", "title": "Event-STVG: Spatio-Temporal Video Grounding for Long-Form Event Reasoning", "year": "2026", "authors": [ "Yizhe Zhang", "Xujia Wang", "Songqi Tang", "Xujie Tang", "Leo Jiahao Jiang", "Jiacheng Wang", "Bowen Qu", "JinBo Li", "Qin Wang", "Ji Qi", "Lei Hou", "Bin Xu", "Juanzi Li" ], "categories": [ "cs.CV" ], "pdf_url": "https://arxiv.org/pdf/2610.22444v1", "abstract": "Spatio-Temporal Video Grounding (STVG) aims to jointly localize a target in space and time given a natural-language query. Despite continued advances, existing STVG benchmarks remain largely centered on localizing targets through instance-specific descriptions, leaving them largely decoupled from long-form reasoning increasingly emphasized in multimodal large language models (MLLMs). In this paper, we introduce Event-STVG, a task that shifts STVG from locating what is described toward event-level reasoning by unifying long-form context, repeated-event reasoning, event localization, role inference, and multi-target grounding. To facilitate research on Event-STVG, we construct a large-scale benchmark comprising 4,016 annotated event occurrences across soccer and seven additional domains. Experiments with state-of-the-art MLLMs reveal substantial difficulty, with the strongest model achieving only 5.4% m-vIoU at the 10-minute horizon. Across domains, we consistently observe confusion among repeated same-type events, spatial grounding bottlenecks, and rare success in jointly grounding all required roles. Event-STVG establishes a new challenge for visual understanding in MLLMs, calling for future models that can jointly reason about complex events across space and time. Code and annotations are available at https://github.com/THU-KEG/EventSTVG" }, "2609.11224.pdf": { "id": "2609.11224v1", "title": "AI Soccer Analyst: Stage-Aware and Verifiable Human-AI Collaboration for Soccer Data Analysis", "year": "2026", "authors": [ "Calvin Yeung", "Keisuke Fujii" ], "categories": [ "cs.HC", "cs.AI" ], "pdf_url": "https://arxiv.org/pdf/2609.11224v1", "abstract": "Sports data analysts translate domain questions into insights by combining computation with sport-specific domain expertise. Large language models ease programming, but prompt-to-report workflows may obscure decisions and evidence. We present AI Soccer Analyst, a mixed-initiative system with revisable stages: Data Understanding, Problem Definition, Structured Planning, Execution, Evidence-Grounded Reporting, and Interaction and Refinement. A formative study with five analysts first informed design goals for automation, verifiability, human control, and accessibility. Subsequently, a task-based evaluation with 16 participants combined system logs, retained artifacts, ratings, and open responses; 33 of 48 tasks met the operational completion criteria. Exploratory tests supported favorable participant perceptions of completed-task output quality, task achievement, reliability, and verifiability after Holm correction. Interaction records showed domain knowledge emerging through clarification, planning, and refinement. These findings position stage-aware human-AI collaboration as a practical approach for producing inspectable, revisable, and verifiable analyses while retaining domain-expert involvement in consequential decisions." }, "2607.14616.pdf": { "id": "2607.14616v5", "title": "SportD: How do VLMs physically strategize?", "year": "2026", "authors": [ "Jasin Cekinmez", "Addison J. Wu", "Haotian Xia", "Kyumin Andrew Shim", "Anay Putty", "Jinglin Xiao", "Zhuohan Liu", "Leo Liu", "Weining Shen" ], "categories": [ "cs.AI", "cs.CV" ], "pdf_url": "https://arxiv.org/pdf/2607.14616v5", "abstract": "Vision-language models (VLMs) can describe a scene, but can they act well within one? We study whether VLMs can make sound strategic decisions, using soccer as an objective testbed with quantifiably-valued actions. We introduce SportD, a dataset and evaluation consisting of 1421 decision scenarios across professional men's and women's soccer games, where a VLM must decide what action to take next. Models on average select the optimal action around 27% of the time, less often than the professional players, and capture markedly less of the value at stake. Furthermore, they exhibit a clear preference for safer actions, favoring lower-variance, lower-value choices that also make less physical progress toward goal. Frontier VLMs are better at estimating whether an action will succeed, placing the highest-success-probability action among their top choices in 72-85% of cases. Yet VLMs systematically conflate likelihood with value, assigning higher value to actions that are more likely to succeed ($\u03c1=+0.30$ to $+0.52$), despite no such relationship in the ground truth ($\u03c1=-0.08$). Modifying the deliberation instructions to encourage risk-taking brings the frontier models closer to the players' skill levels. SportD opens a new direction for rigorously evaluating physical strategic decision-making in VLMs, showing that careful decomposition of their choices can reveal the mechanisms underlying systematic biases such as risk aversion." }, "2606.09452.pdf": { "id": "2606.09452v2", "title": "Quantifying defensive pressure on the ball carrier in soccer based on minimum arrival time", "year": "2026", "authors": [ "Takuma Narizuka", "Ikuya Sakamoto", "Ken Yamamoto", "Yoshihiro Yamazaki" ], "categories": [ "physics.soc-ph" ], "pdf_url": "https://arxiv.org/pdf/2606.09452v2", "abstract": "Defensive pressure on the ball carrier is a fundamental component of soccer tactics. Existing pressure measures often involve additional modeling assumptions, which may reduce interpretability. In this study, we quantify defensive pressure as the opponent minimum arrival time to the ball-carrier location, computed from a physics-based motion model. Using synchronized event and tracking data from all 306 matches in the top division of the Japan Professional Football League during the 2023 season, we analyze the statistical characteristics and temporal evolution of this quantity during ball-possession intervals. The results show that the opponent minimum arrival time tends to decrease during possession and to increase again at the start of the next possession after ball release. We also find that possessions starting under stronger defensive pressure tend to yield smaller ball progression, and that, for intentional open-play passes, possessions ending under stronger pressure are more likely to be lost. These findings indicate that minimum arrival time provides an interpretable and physically grounded measure of immediate defensive pressure on the ball carrier. The proposed framework provides a simple and interpretable baseline for quantifying pressing dynamics from tracking data." }, "2605.15650.pdf": { "id": "2605.15650v1", "title": "MyoChallenge 2025: A New Benchmark for Human Athletic Intelligence", "year": "2026", "authors": [ "Cheryl Wang", "Chun Kwang Tan", "Balint K. Hodossy", "Eric Lyu", "Jun Guo", "Wentao Zhao", "Huaping Liu", "Chengkun Li", "Merkourios Simos", "Bianca Ziliotto", "Alexander Mathis", "Siyuan Liu", "Jiahao Chen", "Shanlin Zhong", "Bo Jiang", "Ci Song", "Yaoye Zhu", "Chenhui Zuo", "Yanan Sui", "Mohamed Irfan Refai", "Massimo Sartori", "Guillaume Durandau", "Vikash Kumar", "Vittorio Caggiano" ], "categories": [ "cs.RO" ], "pdf_url": "https://arxiv.org/pdf/2605.15650v1", "abstract": "Athletic performance represents the pinnacle of human motor intelligence, demanding rapid choices, precise control, agility, and coordinated physical execution. Replicating this seamless combination of capabilities remains elusive in current artificial intelligence and robotic systems. Concurrently, understanding the biological mastery of these movements is hindered because complex muscle coordination is rarely measured in vivo due to the limitations of physical equipment. To bridge this fundamental gap in understanding, MyoChallenge at NeurIPS 2025 established a pioneering benchmark for motor control intelligence in sports, leveraging high-fidelity musculoskeletal models within physics simulation combined with machine learning-driven algorithms. The competition introduces two distinct tracks emphasizing either upper or lower limbs control: a table tennis rally task utilizing a biomechanic upper limb composed of an arm with a hand and a trunk; and a soccer penalty kick using a biomechanic model of legs and a trunk. Marking the fourth iteration of the MyoChallenge series, this event attracted almost 70 teams and over 560 submissions globally, uniting a diverse community ranging from physicians and neuroscientists to machine learning experts. The competition facilitated the development of several state-of-the-art control algorithms for a musculoskeletal system capable of sports agility, leveraging techniques such as physics-based motion planners, on-policy behaviour cloning, hierarchical planning, and muscle synergies. By integrating standardized tasks and physiologically realistic models into the open-source framework of MyoSuite, MyoChallenge'25 serves as a reproducible and reusable testbed to accelerate interdisciplinary research across machine learning, biomechanics, sports science, and neuroscience. Project page: https://www.myosuite.org//myochallenge/myochallenge-2025." }, "2604.04947.pdf": { "id": "2604.04947v1", "title": "SUMMIR: A Hallucination-Aware Framework for Ranking Sports Insights from LLMs", "year": "2026", "authors": [ "Nitish Kumar", "Sannu Kumar", "S Akash", "Manish Gupta", "Ankith Karat", "Sriparna Saha" ], "categories": [ "cs.IR", "cs.AI" ], "pdf_url": "https://arxiv.org/pdf/2604.04947v1", "abstract": "With the rapid proliferation of online sports journalism, extracting meaningful pre-game and post-game insights from articles is essential for enhancing user engagement and comprehension. In this paper, we address the task of automatically extracting such insights from articles published before and after matches. We curate a dataset of 7,900 news articles covering 800 matches across four major sports: Cricket, Soccer, Basketball, and Baseball. To ensure contextual relevance, we employ a two-step validation pipeline leveraging both open-source and proprietary large language models (LLMs). We then utilize multiple state-of-the-art LLMs (GPT-4o, Qwen2.5-72B-Instruct, Llama-3.3-70B-Instruct, and Mixtral-8x7B-Instruct-v0.1) to generate comprehensive insights. The factual accuracy of these outputs is rigorously assessed using a FactScore-based methodology, complemented by hallucination detection via the SummaC (Summary Consistency) framework with GPT-4o. Finally, we propose SUMMIR (Sentence Unified Multimetric Model for Importance Ranking), a novel architecture designed to rank insights based on user-specific interests. Our results demonstrate the effectiveness of this approach in generating high-quality, relevant insights, while also revealing significant differences in factual consistency and interestingness across LLMs. This work contributes a robust framework for automated, reliable insight generation from sports news content. The source code is availble here https://github.com/nitish-iitp/SUMMIR." }, "2512.08500.pdf": { "id": "2512.08500v1", "title": "Learning to Control Physically-simulated 3D Characters via Generating and Mimicking 2D Motions", "year": "2025", "authors": [ "Jianan Li", "Xiao Chen", "Tao Huang", "Tien-Tsin Wong" ], "categories": [ "cs.GR", "cs.CV" ], "pdf_url": "https://arxiv.org/pdf/2512.08500v1", "abstract": "Video data is more cost-effective than motion capture data for learning 3D character motion controllers, yet synthesizing realistic and diverse behaviors directly from videos remains challenging. Previous approaches typically rely on off-the-shelf motion reconstruction techniques to obtain 3D trajectories for physics-based imitation. These reconstruction methods struggle with generalizability, as they either require 3D training data (potentially scarce) or fail to produce physically plausible poses, hindering their application to challenging scenarios like human-object interaction (HOI) or non-human characters. We tackle this challenge by introducing Mimic2DM, a novel motion imitation framework that learns the control policy directly and solely from widely available 2D keypoint trajectories extracted from videos. By minimizing the reprojection error, we train a general single-view 2D motion tracking policy capable of following arbitrary 2D reference motions in physics simulation, using only 2D motion data. The policy, when trained on diverse 2D motions captured from different or slightly different viewpoints, can further acquire 3D motion tracking capabilities by aggregating multiple views. Moreover, we develop a transformer-based autoregressive 2D motion generator and integrate it into a hierarchical control framework, where the generator produces high-quality 2D reference trajectories to guide the tracking policy. We show that the proposed approach is versatile and can effectively learn to synthesize physically plausible and diverse motions across a range of domains, including dancing, soccer dribbling, and animal movements, without any reliance on explicit 3D motion data. Project Website: https://jiann-li.github.io/mimic2dm/" }, "2512.06571.pdf": { "id": "2512.06571v3", "title": "Learning Agile Striker Skills for Humanoid Soccer Robots from Noisy Sensory Input", "year": "2025", "authors": [ "Zifan Xu", "Myoungkyu Seo", "Dongmyeong Lee", "Hao Fu", "Jiaheng Hu", "Jiaxun Cui", "Yuqian Jiang", "Zhihan Wang", "Anastasiia Brund", "Joydeep Biswas", "Peter Stone" ], "categories": [ "cs.RO" ], "pdf_url": "https://arxiv.org/pdf/2512.06571v3", "abstract": "Learning fast and robust ball-kicking skills is a critical capability for humanoid soccer robots, yet it remains a challenging problem due to the need for rapid leg swings, postural stability on a single support foot, and robustness under noisy sensory input and external perturbations (e.g., opponents). This paper presents a reinforcement learning (RL)-based system that enables humanoid robots to execute robust continual ball-kicking with adaptability to different ball-goal configurations. The system extends a typical teacher-student training framework -- in which a \"teacher\" policy is trained with ground truth state information and the \"student\" learns to mimic it with noisy, imperfect sensing -- by including four training stages: (1) long-distance ball chasing (teacher); (2) directional kicking (teacher); (3) teacher policy distillation (student); and (4) student adaptation and refinement (student). Key design elements -- including tailored reward functions, realistic noise modeling, and online constrained RL for adaptation and refinement -- are critical for closing the sim-to-real gap and sustaining performance under perceptual uncertainty. Extensive evaluations in both simulation and on a real robot demonstrate strong kicking accuracy and goal-scoring success across diverse ball-goal configurations. Ablation studies further highlight the necessity of the constrained RL, noise modeling, and the adaptation stage. This work presents a system for learning robust continual humanoid ball-kicking under imperfect perception, establishing a benchmark task for visuomotor skill learning in humanoid whole-body control." }, "2508.17611.pdf": { "id": "2508.17611v1", "title": "Evaluating Movement Initiation Timing in Ultimate Frisbee via Temporal Counterfactuals", "year": "2025", "authors": [ "Shunsuke Iwashita", "Ning Ding", "Keisuke Fujii" ], "categories": [ "cs.AI" ], "pdf_url": "https://arxiv.org/pdf/2508.17611v1", "abstract": "Ultimate is a sport where points are scored by passing a disc and catching it in the opposing team's end zone. In Ultimate, the player holding the disc cannot move, making field dynamics primarily driven by other players' movements. However, current literature in team sports has ignored quantitative evaluations of when players initiate such unlabeled movements in game situations. In this paper, we propose a quantitative evaluation method for movement initiation timing in Ultimate Frisbee. First, game footage was recorded using a drone camera, and players' positional data was obtained, which will be published as UltimateTrack dataset. Next, players' movement initiations were detected, and temporal counterfactual scenarios were generated by shifting the timing of movements using rule-based approaches. These scenarios were analyzed using a space evaluation metric based on soccer's pitch control reflecting the unique rules of Ultimate. By comparing the spatial evaluation values across scenarios, the difference between actual play and the most favorable counterfactual scenario was used to quantitatively assess the impact of movement timing. We validated our method and show that sequences in which the disc was actually thrown to the receiver received higher evaluation scores than the sequences without a throw. In practical verifications, the higher-skill group displays a broader distribution of time offsets from the model's optimal initiation point. These findings demonstrate that the proposed metric provides an objective means of assessing movement initiation timing, which has been difficult to quantify in unlabeled team sport plays." }, "2508.01802.pdf": { "id": "2508.01802v2", "title": "SoccerTrack v2: A Full-Pitch Panoramic Video Dataset for Game State Reconstruction and Ball Action Spotting", "year": "2025", "authors": [ "Atom Scott", "Ikuma Uchida", "Kento Kuroda", "Yufi Kim", "Keisuke Fujii" ], "categories": [ "cs.CV" ], "pdf_url": "https://arxiv.org/pdf/2508.01802v2", "abstract": "Soccer analytics draws on two kinds of information: spatio-temporal data describing where players and the ball are, and event data describing what they do. Public datasets offer them apart, or together only on broadcast footage that leaves players outside the frame unobserved. SoccerTrack v2 combines continuous full-pitch video, long player trajectories and actor-linked events in one resource: ten university-level matches, 932 minutes of fixed-camera 4K panoramic video, annotated per frame with metric pitch coordinates, jersey numbers and persistent identities, roles and team sides for all players, and with ball action events in twelve classes, linked to the acting players through the same identifiers used in the trajectories. We fix a match-level split and report baselines for two tasks. For game state reconstruction, we run a full pipeline over all twenty halves and find that GS-HOTA scores degrade as sequence length increases. For ball action spotting, we train a model on the player trajectories, with and without the ball track. The data, the split and the evaluation tooling are released so that both tasks can be developed and compared at match length on the same footage." }, "2505.12679.pdf": { "id": "2505.12679v3", "title": "Dribble Master: Learning Agile Humanoid Dribbling through Legged Locomotion", "year": "2025", "authors": [ "Zhuoheng Wang", "Jinyin Zhou", "Qi Wu" ], "categories": [ "cs.RO" ], "pdf_url": "https://arxiv.org/pdf/2505.12679v3", "abstract": "Humanoid soccer dribbling is a highly challenging task that demands dexterous ball manipulation while maintaining dynamic balance. Traditional rule-based methods often struggle to achieve accurate ball control due to their reliance on fixed walking patterns and limited adaptability to real-time ball dynamics. To address these challenges, we propose a two-stage curriculum learning framework that enables a humanoid robot to acquire dribbling skills without explicit dynamics or predefined trajectories. In the first stage, the robot learns basic locomotion skills; in the second stage, we fine-tune the policy for agile dribbling maneuvers. We further introduce a virtual camera model in simulation that simulates the field of view and perception constraints of the real robot, enabling realistic ball perception during training. We also design heuristic rewards to encourage active sensing, promoting a broader visual range for continuous ball perception. The policy is trained in simulation and successfully transferred to a physical humanoid robot. Experiment results demonstrate that our method enables effective ball manipulation, achieving flexible and visually appealing dribbling behaviors across multiple environments. This work highlights the potential of reinforcement learning in developing agile humanoid soccer robots. Additional details and videos are available at https://zhuoheng0910.github.io/dribble-master/." }, "2505.14711.pdf": { "id": "2505.14711v1", "title": "Space evaluation at the starting point of soccer transitions", "year": "2025", "authors": [ "Yohei Ogawa", "Rikuhei Umemoto", "Keisuke Fujii" ], "categories": [ "stat.AP", "cs.AI" ], "pdf_url": "https://arxiv.org/pdf/2505.14711v1", "abstract": "Soccer is a sport played on a pitch where effective use of space is crucial. Decision-making during transitions, when possession switches between teams, has been increasingly important, but research on space evaluation in these moments has been limited. Recent space evaluation methods such as OBSO (Off-Ball Scoring Opportunity) use scoring probability, so it is not well-suited for assessing areas far from the goal, where transitions typically occur. In this paper, we propose OBPV (Off-Ball Positioning Value) to evaluate space across the pitch, including the starting points of transitions. OBPV extends OBSO by introducing the field value model, which evaluates the entire pitch, and by employing the transition kernel model, which reflects positional specificity through kernel density estimation of pass distributions. Experiments using La Liga 2023/24 season tracking and event data show that OBPV highlights effective space utilization during counter-attacks and reveals team-specific characteristics in how the teams utilize space after positive and negative transitions." }, "2503.05092.pdf": { "id": "2503.05092v1", "title": "Multi-Robot Collaboration through Reinforcement Learning and Abstract Simulation", "year": "2025", "authors": [ "Adam Labiosa", "Josiah P. Hanna" ], "categories": [ "cs.RO", "cs.AI" ], "pdf_url": "https://arxiv.org/pdf/2503.05092v1", "abstract": "Teams of people coordinate to perform complex tasks by forming abstract mental models of world and agent dynamics. The use of abstract models contrasts with much recent work in robot learning that uses a high-fidelity simulator and reinforcement learning (RL) to obtain policies for physical robots. Motivated by this difference, we investigate the extent to which so-called abstract simulators can be used for multi-agent reinforcement learning (MARL) and the resulting policies successfully deployed on teams of physical robots. An abstract simulator models the robot's target task at a high-level of abstraction and discards many details of the world that could impact optimal decision-making. Policies are trained in an abstract simulator then transferred to the physical robot by making use of separately-obtained low-level perception and motion control modules. We identify three key categories of modifications to the abstract simulator that enable policy transfer to physical robots: simulation fidelity enhancements, training optimizations and simulation stochasticity. We then run an empirical study with extensive ablations to determine the value of each modification category for enabling policy transfer in cooperative robot soccer tasks. We also compare the performance of policies produced by our method with a well-tuned non-learning-based behavior architecture from the annual RoboCup competition and find that our approach leads to a similar level of performance. Broadly we show that MARL can be use to train cooperative physical robot behaviors using highly abstract models of the world." }, "2412.09417.pdf": { "id": "2412.09417v2", "title": "Reinforcement Learning Within the Classical Robotics Stack: A Case Study in Robot Soccer", "year": "2024", "authors": [ "Adam Labiosa", "Zhihan Wang", "Siddhant Agarwal", "William Cong", "Geethika Hemkumar", "Abhinav Narayan Harish", "Benjamin Hong", "Josh Kelle", "Chen Li", "Yuhao Li", "Zisen Shao", "Peter Stone", "Josiah P. Hanna" ], "categories": [ "cs.RO", "cs.AI", "cs.LG" ], "pdf_url": "https://arxiv.org/pdf/2412.09417v2", "abstract": "Robot decision-making in partially observable, real-time, dynamic, and multi-agent environments remains a difficult and unsolved challenge. Model-free reinforcement learning (RL) is a promising approach to learning decision-making in such domains, however, end-to-end RL in complex environments is often intractable. To address this challenge in the RoboCup Standard Platform League (SPL) domain, we developed a novel architecture integrating RL within a classical robotics stack, while employing a multi-fidelity sim2real approach and decomposing behavior into learned sub-behaviors with heuristic selection. Our architecture led to victory in the 2024 RoboCup SPL Challenge Shield Division. In this work, we fully describe our system's architecture and empirically analyze key design decisions that contributed to its success. Our approach demonstrates how RL-based behaviors can be integrated into complete robot behavior architectures." }, "2411.04942.pdf": { "id": "2411.04942v1", "title": "A Reinforcement Learning-Based Automatic Video Editing Method Using Pre-trained Vision-Language Model", "year": "2024", "authors": [ "Panwen Hu", "Nan Xiao", "Feifei Li", "Yongquan Chen", "Rui Huang" ], "categories": [ "cs.CV" ], "pdf_url": "https://arxiv.org/pdf/2411.04942v1", "abstract": "In this era of videos, automatic video editing techniques attract more and more attention from industry and academia since they can reduce workloads and lower the requirements for human editors. Existing automatic editing systems are mainly scene- or event-specific, e.g., soccer game broadcasting, yet the automatic systems for general editing, e.g., movie or vlog editing which covers various scenes and events, were rarely studied before, and converting the event-driven editing method to a general scene is nontrivial. In this paper, we propose a two-stage scheme for general editing. Firstly, unlike previous works that extract scene-specific features, we leverage the pre-trained Vision-Language Model (VLM) to extract the editing-relevant representations as editing context. Moreover, to close the gap between the professional-looking videos and the automatic productions generated with simple guidelines, we propose a Reinforcement Learning (RL)-based editing framework to formulate the editing problem and train the virtual editor to make better sequential editing decisions. Finally, we evaluate the proposed method on a more general editing task with a real movie dataset. Experimental results demonstrate the effectiveness and benefits of the proposed context representation and the learning ability of our RL-based editing framework." }, "2409.17129.pdf": { "id": "2409.17129v1", "title": "Bayesian Bivariate Conway-Maxwell-Poisson Regression Model for Correlated Count Data in Sports", "year": "2024", "authors": [ "Mauro Florez", "Michele Guindani", "Marina Vannucci" ], "categories": [ "stat.ME", "stat.AP" ], "pdf_url": "https://arxiv.org/pdf/2409.17129v1", "abstract": "Count data play a crucial role in sports analytics, providing valuable insights into various aspects of the game. Models that accurately capture the characteristics of count data are essential for making reliable inferences. In this paper, we propose the use of the Conway-Maxwell-Poisson (CMP) model for analyzing count data in sports. The CMP model offers flexibility in modeling data with different levels of dispersion. Here we consider a bivariate CMP model that models the potential correlation between home and away scores by incorporating a random effect specification. We illustrate the advantages of the CMP model through simulations. We then analyze data from baseball and soccer games before, during, and after the COVID-19 pandemic. The performance of our proposed CMP model matches or outperforms standard Poisson and Negative Binomial models, providing a good fit and an accurate estimation of the observed effects in count data with any level of dispersion. The results highlight the robustness and flexibility of the CMP model in analyzing count data in sports, making it a suitable default choice for modeling a diverse range of count data types in sports, where the data dispersion may vary." }, "2310.18247.pdf": { "id": "2310.18247v3", "title": "Guided Data Augmentation for Offline Reinforcement Learning and Imitation Learning", "year": "2023", "authors": [ "Nicholas E. Corrado", "Yuxiao Qu", "John U. Balis", "Adam Labiosa", "Josiah P. Hanna" ], "categories": [ "cs.LG", "cs.RO" ], "pdf_url": "https://arxiv.org/pdf/2310.18247v3", "abstract": "In offline reinforcement learning (RL), an RL agent learns to solve a task using only a fixed dataset of previously collected data. While offline RL has been successful in learning real-world robot control policies, it typically requires large amounts of expert-quality data to learn effective policies that generalize to out-of-distribution states. Unfortunately, such data is often difficult and expensive to acquire in real-world tasks. Several recent works have leveraged data augmentation (DA) to inexpensively generate additional data, but most DA works apply augmentations in a random fashion and ultimately produce highly suboptimal augmented experience. In this work, we propose Guided Data Augmentation (GuDA), a human-guided DA framework that generates expert-quality augmented data. The key insight behind GuDA is that while it may be difficult to demonstrate the sequence of actions required to produce expert data, a user can often easily characterize when an augmented trajectory segment represents progress toward task completion. Thus, a user can restrict the space of possible augmentations to automatically reject suboptimal augmented data. To extract a policy from GuDA, we use off-the-shelf offline reinforcement learning and behavior cloning algorithms. We evaluate GuDA on a physical robot soccer task as well as simulated D4RL navigation tasks, a simulated autonomous driving task, and a simulated soccer task. Empirically, GuDA enables learning given a small initial dataset of potentially suboptimal experience and outperforms a random DA strategy as well as a model-based DA strategy." }, "2308.08610.pdf": { "id": "2308.08610v1", "title": "FootGPT : A Large Language Model Development Experiment on a Minimal Setting", "year": "2023", "authors": [ "Eren Unlu" ], "categories": [ "cs.CL", "cs.AI" ], "pdf_url": "https://arxiv.org/pdf/2308.08610v1", "abstract": "With recent empirical observations, it has been argued that the most significant aspect of developing accurate language models may be the proper dataset content and training strategy compared to the number of neural parameters, training duration or dataset size. Following this argument, we opted to fine tune a one billion parameter size trained general purpose causal language model with a dataset curated on team statistics of the Italian football league first ten game weeks, using low rank adaptation. The limited training dataset was compiled based on a framework where a powerful commercial large language model provides distilled paragraphs and question answer pairs as intended. The training duration was kept relatively short to provide a basis for our minimal setting exploration. We share our key observations on the process related to developing a specific purpose language model which is intended to interpret soccer data with constrained resources in this article." }, "2301.03462.pdf": { "id": "2301.03462v2", "title": "Unleashing the Power of Shared Label Structures for Human Activity Recognition", "year": "2023", "authors": [ "Xiyuan Zhang", "Ranak Roy Chowdhury", "Jiayun Zhang", "Dezhi Hong", "Rajesh K. Gupta", "Jingbo Shang" ], "categories": [ "cs.LG", "cs.AI", "eess.SP" ], "pdf_url": "https://arxiv.org/pdf/2301.03462v2", "abstract": "Current human activity recognition (HAR) techniques regard activity labels as integer class IDs without explicitly modeling the semantics of class labels. We observe that different activity names often have shared structures. For example, \"open door\" and \"open fridge\" both have \"open\" as the action; \"kicking soccer ball\" and \"playing tennis ball\" both have \"ball\" as the object. Such shared structures in label names can be translated to the similarity in sensory data and modeling common structures would help uncover knowledge across different activities, especially for activities with limited samples. In this paper, we propose SHARE, a HAR framework that takes into account shared structures of label names for different activities. To exploit the shared structures, SHARE comprises an encoder for extracting features from input sensory time series and a decoder for generating label names as a token sequence. We also propose three label augmentation techniques to help the model more effectively capture semantic structures across activities, including a basic token-level augmentation, and two enhanced embedding-level and sequence-level augmentations utilizing the capabilities of pre-trained models. SHARE outperforms state-of-the-art HAR models in extensive experiments on seven HAR benchmark datasets. We also evaluate in few-shot learning and label imbalance settings and observe even more significant performance gap." }, "2212.06504.pdf": { "id": "2212.06504v1", "title": "Accelerated structured matrix factorization", "year": "2022", "authors": [ "Lorenzo Schiavon", "Bernardo Nipoti", "Antonio Canale" ], "categories": [ "stat.ME", "stat.ML" ], "pdf_url": "https://arxiv.org/pdf/2212.06504v1", "abstract": "Matrix factorization exploits the idea that, in complex high-dimensional data, the actual signal typically lies in lower-dimensional structures. These lower dimensional objects provide useful insight, with interpretability favored by sparse structures. Sparsity, in addition, is beneficial in terms of regularization and, thus, to avoid over-fitting. By exploiting Bayesian shrinkage priors, we devise a computationally convenient approach for high-dimensional matrix factorization. The dependence between row and column entities is modeled by inducing flexible sparse patterns within factors. The availability of external information is accounted for in such a way that structures are allowed while not imposed. Inspired by boosting algorithms, we pair the the proposed approach with a numerical strategy relying on a sequential inclusion and estimation of low-rank contributions, with data-driven stopping rule. Practical advantages of the proposed approach are demonstrated by means of a simulation study and the analysis of soccer heatmaps obtained from new generation tracking data." }, "2211.08532.pdf": { "id": "2211.08532v1", "title": "Omnidirectional robot modeling and simulation", "year": "2022", "authors": [ "Sandro Costa Magalh\u00e3es", "Ant\u00f3nio Paulo Moreira", "Paulo Costa" ], "categories": [ "cs.RO", "eess.SY" ], "pdf_url": "https://arxiv.org/pdf/2211.08532v1", "abstract": "A robot simulation system is a basic need for any robotics application. With it, developers' teams of robots can test their algorithms and make initial calibrations without risk of damage to the real robots, assuring safety. However, building these simulation environments is usually time-consuming work, and when considering robot fleets, the simulation reveals to be computing expensive. With it, developers building teams of robots can test their algorithms and make initial calibrations without risk of damage to the real robots, assuring safety. An omnidirectional robot from the 5DPO robotics soccer team served to test this approach. The modeling issue was divided into two steps: modeling the motor's non-linear features and modeling the general behavior of the robot. A proper fitting of the robot was reached, considering the velocity robot's response." }, "2209.11252.pdf": { "id": "2209.11252v1", "title": "XF2T: Cross-lingual Fact-to-Text Generation for Low-Resource Languages", "year": "2022", "authors": [ "Shivprasad Sagare", "Tushar Abhishek", "Bhavyajeet Singh", "Anubhav Sharma", "Manish Gupta", "Vasudeva Varma" ], "categories": [ "cs.CL" ], "pdf_url": "https://arxiv.org/pdf/2209.11252v1", "abstract": "Multiple business scenarios require an automated generation of descriptive human-readable text from structured input data. Hence, fact-to-text generation systems have been developed for various downstream tasks like generating soccer reports, weather and financial reports, medical reports, person biographies, etc. Unfortunately, previous work on fact-to-text (F2T) generation has focused primarily on English mainly due to the high availability of relevant datasets. Only recently, the problem of cross-lingual fact-to-text (XF2T) was proposed for generation across multiple languages alongwith a dataset, XALIGN for eight languages. However, there has been no rigorous work on the actual XF2T generation problem. We extend XALIGN dataset with annotated data for four more languages: Punjabi, Malayalam, Assamese and Oriya. We conduct an extensive study using popular Transformer-based text generation models on our extended multi-lingual dataset, which we call XALIGNV2. Further, we investigate the performance of different text generation strategies: multiple variations of pretraining, fact-aware embeddings and structure-aware input encoding. Our extensive experiments show that a multi-lingual mT5 model which uses fact-aware embeddings with structure-aware input encoding leads to best results on average across the twelve languages. We make our code, dataset and model publicly available, and hope that this will help advance further research in this critical area." }, "2011.13851.pdf": { "id": "2011.13851v1", "title": "Real-time Active Vision for a Humanoid Soccer Robot Using Deep Reinforcement Learning", "year": "2020", "authors": [ "Soheil Khatibi", "Meisam Teimouri", "Mahdi Rezaei" ], "categories": [ "cs.RO", "cs.AI", "cs.CV", "cs.LG", "eess.IV" ], "pdf_url": "https://arxiv.org/pdf/2011.13851v1", "abstract": "In this paper, we present an active vision method using a deep reinforcement learning approach for a humanoid soccer-playing robot. The proposed method adaptively optimises the viewpoint of the robot to acquire the most useful landmarks for self-localisation while keeping the ball into its viewpoint. Active vision is critical for humanoid decision-maker robots with a limited field of view. To deal with an active vision problem, several probabilistic entropy-based approaches have previously been proposed which are highly dependent on the accuracy of the self-localisation model. However, in this research, we formulate the problem as an episodic reinforcement learning problem and employ a Deep Q-learning method to solve it. The proposed network only requires the raw images of the camera to move the robot's head toward the best viewpoint. The model shows a very competitive rate of 80% success rate in achieving the best viewpoint. We implemented the proposed method on a humanoid robot simulated in Webots simulator. Our evaluations and experimental results show that the proposed method outperforms the entropy-based methods in the RoboCup context, in cases with high self-localisation errors." }, "2004.07209.pdf": { "id": "2004.07209v1", "title": "Using Player's Body-Orientation to Model Pass Feasibility in Soccer", "year": "2020", "authors": [ "Adri\u00e0 Arbu\u00e9s-Sang\u00fcesa", "Adri\u00e1n Mart\u00edn", "Javier Fern\u00e1ndez", "Coloma Ballester", "Gloria Haro" ], "categories": [ "cs.CV" ], "pdf_url": "https://arxiv.org/pdf/2004.07209v1", "abstract": "Given a monocular video of a soccer match, this paper presents a computational model to estimate the most feasible pass at any given time. The method leverages offensive player's orientation (plus their location) and opponents' spatial configuration to compute the feasibility of pass events within players of the same team. Orientation data is gathered from body pose estimations that are properly projected onto the 2D game field; moreover, a geometrical solution is provided, through the definition of a feasibility measure, to determine which players are better oriented towards each other. Once analyzed more than 6000 pass events, results show that, by including orientation as a feasibility measure, a robust computational model can be built, reaching more than 0.7 Top-3 accuracy. Finally, the combination of the orientation feasibility measure with the recently introduced Expected Possession Value metric is studied; promising results are obtained, thus showing that existing models can be refined by using orientation as a key feature. These models could help both coaches and analysts to have a better understanding of the game and to improve the players' decision-making process." }, "1912.05445.pdf": { "id": "1912.05445v2", "title": "FootAndBall: Integrated player and ball detector", "year": "2019", "authors": [ "Jacek Komorowski", "Grzegorz Kurzejamski", "Grzegorz Sarwas" ], "categories": [ "cs.CV" ], "pdf_url": "https://arxiv.org/pdf/1912.05445v2", "abstract": "The paper describes a deep neural network-based detector dedicated for ball and players detection in high resolution, long shot, video recordings of soccer matches. The detector, dubbed FootAndBall, has an efficient fully convolutional architecture and can operate on input video stream with an arbitrary resolution. It produces ball confidence map encoding the position of the detected ball, player confidence map and player bounding boxes tensor encoding players' positions and bounding boxes. The network uses Feature Pyramid Network desing pattern, where lower level features with higher spatial resolution are combined with higher level features with bigger receptive field. This improves discriminability of small objects (the ball) as larger visual context around the object of interest is taken into account for the classification. Due to its specialized design, the network has two orders of magnitude less parameters than a generic deep neural network-based object detector, such as SSD or YOLO. This allows real-time processing of high resolution input video stream. Our code and pre-trained model can be found on the project website: https://github.com/jac99/FootAndBall ." }, "1910.10620.pdf": { "id": "1910.10620v1", "title": "Learning Humanoid Robot Running Skills through Proximal Policy Optimization", "year": "2019", "authors": [ "Luckeciano C. Melo", "Marcos R. O. A. Maximo" ], "categories": [ "cs.RO", "cs.AI", "cs.LG" ], "pdf_url": "https://arxiv.org/pdf/1910.10620v1", "abstract": "In the current level of evolution of Soccer 3D, motion control is a key factor in team's performance. Recent works takes advantages of model-free approaches based on Machine Learning to exploit robot dynamics in order to obtain faster locomotion skills, achieving running policies and, therefore, opening a new research direction in the Soccer 3D environment. In this work, we present a methodology based on Deep Reinforcement Learning that learns running skills without any prior knowledge, using a neural network whose inputs are related to robot's dynamics. Our results outperformed the previous state-of-the-art sprint velocity reported in Soccer 3D literature by a significant margin. It also demonstrated improvement in sample efficiency, being able to learn how to run in just few hours. We reported our results analyzing the training procedure and also evaluating the policies in terms of speed, reliability and human similarity. Finally, we presented key factors that lead us to improve previous results and shared some ideas for future work." }, "1810.10658.pdf": { "id": "1810.10658v1", "title": "Sports Camera Calibration via Synthetic Data", "year": "2018", "authors": [ "Jianhui Chen", "James J. Little" ], "categories": [ "cs.CV" ], "pdf_url": "https://arxiv.org/pdf/1810.10658v1", "abstract": "Calibrating sports cameras is important for autonomous broadcasting and sports analysis. Here we propose a highly automatic method for calibrating sports cameras from a single image using synthetic data. First, we develop a novel camera pose engine. The camera pose engine has only three significant free parameters so that it can effectively generate a lot of camera poses and corresponding edge (i.e, field marking) images. Then, we learn compact deep features via a siamese network from paired edge image and camera pose and build a feature-pose database. After that, we use a novel two-GAN (generative adversarial network) model to detect field markings in real images. Finally, we query an initial camera pose from the feature-pose database and refine camera poses using truncated distance images. We evaluate our method on both synthetic and real data. Our method not only demonstrates the robustness on the synthetic data but also achieves the state-of-the-art accuracy on a standard soccer dataset and very high performance on a volleyball dataset." }, "1807.09205.pdf": { "id": "1807.09205v1", "title": "End-to-End Deep Imitation Learning: Robot Soccer Case Study", "year": "2018", "authors": [ "Okan A\u015f\u0131k", "Binnur G\u00f6rer", "H. Levent Ak\u0131n" ], "categories": [ "cs.RO", "cs.AI" ], "pdf_url": "https://arxiv.org/pdf/1807.09205v1", "abstract": "In imitation learning, behavior learning is generally done using the features extracted from the demonstration data. Recent deep learning algorithms enable the development of machine learning methods that can get high dimensional data as an input. In this work, we use imitation learning to teach the robot to dribble the ball to the goal. We use B-Human robot software to collect demonstration data and a deep convolutional network to represent the policies. We use top and bottom camera images of the robot as input and speed commands as outputs. The CNN policy learns the mapping between the series of images and speed commands. In 3D realistic robotics simulator experiments, we show that the robot is able to learn to search the ball and dribble the ball, but it struggles to align to the goal. The best-proposed policy model learns to score 4 goals out of 20 test episodes." }, "1711.07832.pdf": { "id": "1711.07832v1", "title": "Situationally Aware Options", "year": "2017", "authors": [ "Daniel J. Mankowitz", "Aviv Tamar", "Shie Mannor" ], "categories": [ "cs.AI" ], "pdf_url": "https://arxiv.org/pdf/1711.07832v1", "abstract": "Hierarchical abstractions, also known as options -- a type of temporally extended action (Sutton et. al. 1999) that enables a reinforcement learning agent to plan at a higher level, abstracting away from the lower-level details. In this work, we learn reusable options whose parameters can vary, encouraging different behaviors, based on the current situation. In principle, these behaviors can include vigor, defence or even risk-averseness. These are some examples of what we refer to in the broader context as Situational Awareness (SA). We incorporate SA, in the form of vigor, into hierarchical RL by defining and learning situationally aware options in a Probabilistic Goal Semi-Markov Decision Process (PG-SMDP). This is achieved using our Situationally Aware oPtions (SAP) policy gradient algorithm which comes with a theoretical convergence guarantee. We learn reusable options in different scenarios in a RoboCup soccer domain (i.e., winning/losing). These options learn to execute with different levels of vigor resulting in human-like behaviours such as `time-wasting' in the winning scenario. We show the potential of the agent to exit bad local optima using reusable options in RoboCup. Finally, using SAP, the agent mitigates feature-based model misspecification in a Bottomless Pit of Death domain." }, "1706.04315.pdf": { "id": "1706.04315v1", "title": "RoboCup 2D Soccer Simulation League: Evaluation Challenges", "year": "2017", "authors": [ "Mikhail Prokopenko", "Peter Wang", "Sebastian Marian", "Aijun Bai", "Xiao Li", "Xiaoping Chen" ], "categories": [ "cs.MA" ], "pdf_url": "https://arxiv.org/pdf/1706.04315v1", "abstract": "We summarise the results of RoboCup 2D Soccer Simulation League in 2016 (Leipzig), including the main competition and the evaluation round. The evaluation round held in Leipzig confirmed the strength of RoboCup-2015 champion (WrightEagle, i.e. WE2015) in the League, with only eventual finalists of 2016 competition capable of defeating WE2015. An extended, post-Leipzig, round-robin tournament which included the top 8 teams of 2016, as well as WE2015, with over 1000 games played for each pair, placed WE2015 third behind the champion team (Gliders2016) and the runner-up (HELIOS2016). This establishes WE2015 as a stable benchmark for the 2D Simulation League. We then contrast two ranking methods and suggest two options for future evaluation challenges. The first one, \"The Champions Simulation League\", is proposed to include 6 previous champions, directly competing against each other in a round-robin tournament, with the view to systematically trace the advancements in the League. The second proposal, \"The Global Challenge\", is aimed to increase the realism of the environmental conditions during the simulated games, by simulating specific features of different participating countries." }, "2101.02104.pdf": { "id": "2101.02104v1", "title": "A Probabilistic Model for Predicting Shot Success in Football", "year": "2021", "authors": [ "Edward Wheatcroft", "Ewelina Sienkiewicz" ], "categories": [ "stat.AP" ], "pdf_url": "https://arxiv.org/pdf/2101.02104v1", "abstract": "Football forecasting models traditionally rate teams on past match results, that is based on the number of goals scored. Goals, however, involve a high element of chance and thus past results often do not reflect the performances of the teams. In recent years, it has become increasingly clear that accounting for other match events such as shots at goal can provide a better indication of the relative strengths of two teams than the number of goals scored. Forecast models based on this information have been shown to be successful in outperforming those based purely on match results. A notable weakness, however, is that this approach does not take into account differences in the probability of shot success among teams. A team that is more likely to score from a shot will need fewer shots to win a match, on average. In this paper, we propose a simple parametric model to predict the probability of a team scoring, given it has taken a shot at goal. We show that the resulting forecasts are able to outperform a model assuming an equal probability of shot success among all teams. We then show that the model can be combined with predictions of the number of shots achieved by each team, and can increase the skill of forecasts of both the match outcome and of whether the total number of goals in a match will exceed 2.5. We assess the performance of the forecasts alongside two betting strategies and find mixed evidence for improved performance." }, "1610.02847.pdf": { "id": "1610.02847v1", "title": "Situational Awareness by Risk-Conscious Skills", "year": "2016", "authors": [ "Daniel J. Mankowitz", "Aviv Tamar", "Shie Mannor" ], "categories": [ "cs.AI" ], "pdf_url": "https://arxiv.org/pdf/1610.02847v1", "abstract": "Hierarchical Reinforcement Learning has been previously shown to speed up the convergence rate of RL planning algorithms as well as mitigate feature-based model misspecification (Mankowitz et. al. 2016a,b, Bacon 2015). To do so, it utilizes hierarchical abstractions, also known as skills -- a type of temporally extended action (Sutton et. al. 1999) to plan at a higher level, abstracting away from the lower-level details. We incorporate risk sensitivity, also referred to as Situational Awareness (SA), into hierarchical RL for the first time by defining and learning risk aware skills in a Probabilistic Goal Semi-Markov Decision Process (PG-SMDP). This is achieved using our novel Situational Awareness by Risk-Conscious Skills (SARiCoS) algorithm which comes with a theoretical convergence guarantee. We show in a RoboCup soccer domain that the learned risk aware skills exhibit complex human behaviors such as `time-wasting' in a soccer game. In addition, the learned risk aware skills are able to mitigate reward-based model misspecification." }, "1604.03614.pdf": { "id": "1604.03614v5", "title": "The Market for English Premier League (EPL) Odds", "year": "2016", "authors": [ "Guanhao Feng", "Nicholas G. Polson", "Jianeng Xu" ], "categories": [ "stat.AP", "stat.ME" ], "pdf_url": "https://arxiv.org/pdf/1604.03614v5", "abstract": "This paper employs a Skellam process to represent real-time betting odds for English Premier League (EPL) soccer games. Given a matrix of market odds on all possible score outcomes, we estimate the expected scoring rates for each team. The expected scoring rates then define the implied volatility of an EPL game. As events in the game evolve, we re-estimate the expected scoring rates and our implied volatility measure to provide a dynamic representation of the market's expectation of the game outcome. Using a dataset of 1520 EPL games from 2012-2016, we show how our model calibrates well to the game outcome. We illustrate our methodology on real-time market odds data for a game between Everton and West Ham in the 2015-2016 season. We show how the implied volatility for the outcome evolves as goals, red cards, and corner kicks occur. Finally, we conclude with directions for future research." }, "2010.11187.pdf": { "id": "2010.11187v3", "title": "G-Elo: Generalization of the Elo algorithm by modelling the discretized margin of victory", "year": "2020", "authors": [ "Leszek Szczecinski" ], "categories": [ "stat.ME", "cs.LG", "math.OC" ], "pdf_url": "https://arxiv.org/pdf/2010.11187v3", "abstract": "In this work we develop a new algorithm for rating of teams (or players) in one-on-one games by exploiting the observed difference of the game-points (such as goals), also known as a margin of victory (MOV). Our objective is to obtain the Elo-style algorithm whose operation is simple to implement and to understand intuitively. This is done in three steps: first, we define the probabilistic model between the teams' skills and the discretized MOV variable: this generalizes the model underpinning the Elo algorithm, where the MOV variable is discretized into three categories (win/loss/draw). Second, with the formal probabilistic model at hand, the optimization required by the maximum likelihood rule is implemented via stochastic gradient; this yields simple on-line equations for the rating updates which are identical in their general form to those characteristic of the Elo algorithm: the main difference lies in the way the scores and the expected scores are defined. Third, we propose a simple method to estimate the coefficients of the model, and thus define the operation of the algorithm; it is done in a closed form using the historical data so the algorithm is tailored to the sport of interest and the coefficients defining its operation are determined in entirely transparent manner. The alternative, optimization-based strategy to find the coefficients is also presented. We show numerical examples based on the results of the association football of the English Premier League and the American football of the National Football League." }, "2609.18207.pdf": { "id": "2609.18207v1", "title": "Reinforcement Learning for Real-Time Vision-Language-Action Policies", "year": "2026", "authors": [ "Perry Dong", "Kuo-Han Hung", "Dorsa Sadigh", "Chelsea Finn" ], "categories": [ "cs.RO", "cs.LG" ], "pdf_url": "https://arxiv.org/pdf/2609.18207v1", "abstract": "Reinforcement learning fine-tuning on top of large, pretrained Vision-Language-Action (VLA) models offers promise for highly reliable robot deployment. However, because of their scale, modern VLA models suffer from high inference latency, so the observation used to select an action is often stale by execution time, creating a distribution shift that can substantially degrade reliability and performance. Prior work has explored asynchronous policy execution to reduce the effect of latency, but these methods are mostly built on imitation learning and offer no mechanism for moving beyond the training distribution toward higher reliability. We close this gap by enabling RL fine-tuning that meets the real-time control requirements of dynamic real-world manipulation. Our approach builds on EXPO-FT, a framework for sample-efficient, reliable VLA fine-tuning with reinforcement learning, and decouples slow, expressive action generation from fast, reactive action edits: a large pretrained VLA proposes action chunks using its strong behavior prior, while a lightweight edit policy performs fast, reactive decision-making by editing actions in response to changes in state, conditioned on the latest observation. We instantiate this as Real-Time EXPO-FT, an RL framework for finetuning real-time VLA policies. On the Kinetix benchmark, Real-Time EXPO-FT enables a delayed policy to achieve the best performance among delayed and non-delayed methods in 10 out of 10 environments. On four dynamic real-world tasks, robot object passing, ball balancing, table soccer kicking, and dynamic object picking, with online robot data capped at 10 minutes, Real-Time EXPO-FT improves average policy performance from 42% to 97%, all without human intervention, demonstrating rapid, sample-efficient adaptation to challenging real-world dynamics. Website: https://pd-perry.github.io/real-time-expo-ft" }, "2608.00876.pdf": { "id": "2608.00876v1", "title": "Who's That Player?: Externalizing Query Interpretation in Spoken XR Sports Interaction", "year": "2026", "authors": [ "Chunggi Lee", "Tica Lin", "Yalong Yang", "Hanspeter Pfister" ], "categories": [ "cs.HC" ], "pdf_url": "https://arxiv.org/pdf/2608.00876v1", "abstract": "XR sports viewing enables spectators to follow play from immersive, spatially anchored perspectives while accessing contextual analytics directly within the scene. In such settings, speech offers a practical interaction modality because text entry and menu navigation can interrupt attention during fast-paced gameplay. However, spoken queries are often underspecified: viewers may omit which player, time period, field location, or metric they intend. When systems resolve these ambiguities implicitly, their assumptions remain hidden, making misinterpretations difficult to notice and correct (repair). We investigate how externalizing a system's interpretation of spoken queries can support inspection and correction of such misunderstandings in XR sports viewing. Through a formative study, we identified four recurring ambiguity types (referential, spatial, temporal, and metric) that characterize ambiguous spoken queries in this context. We develop a design space that organizes externalization along three dimensions (ambiguity type, interpretation state, externalization strategy) and instantiate it in an interactive XR soccer viewing system that combines situated visual cues with supporting analytic views. A within-subjects user study (N=16) comparing externalized interpretation against a voice-only baseline reveals that externalization is associated with higher inspectability on most measured dimensions and increased explicit repair language overall. However, repair occurred in only 38% of misaligned externalization trials, and this visibility-action gap varied by ambiguity type, indicating that transparency and correction affordance are orthogonal design axes." }, "2606.12106.pdf": { "id": "2606.12106v1", "title": "MSUE: Multi-Modal Soccer Understanding Expert", "year": "2026", "authors": [ "Litao Li", "Yibo Yu", "Yufeng Hu", "Zhuo Yang", "Jiali Wen", "Yixin Chen", "Yixi Zhou" ], "categories": [ "cs.CV", "cs.AI" ], "pdf_url": "https://arxiv.org/pdf/2606.12106v1", "abstract": "This paper presents our solution to the 2026 SoccerNet VQA Challenge. We first develop a cost-effective data synthesis pipeline driven by a Vision-Language Model (VLM), which systematically restructures raw domain data into diverse VQA samples, including concise answers and long-form responses. Second, we propose MSUE, a multi-expert question answering architecture that employs a Large Language Model (LLM) to dynamically dispatch questions to text, image, and video experts. These experts are instantiated as a strong text baseline Gemini3-Flash, a fine-tuned Qwen3-VL, and an external knowledge base, respectively, working collaboratively to enhance VQA performance. MSUE achieves an accuracy of \\textbf{0.95} on the challenge benchmark, securing third place in the leaderboard." }, "2604.01990.pdf": { "id": "2604.01990v1", "title": "Evaluating Deep Surrogate Models for Knee Joint Contact Mechanics Under Input-Limited Conditions", "year": "2026", "authors": [ "Zhengye Pan", "Jianwei Zuo", "Jiajia Luo" ], "categories": [ "q-bio.QM" ], "pdf_url": "https://arxiv.org/pdf/2604.01990v1", "abstract": "Background and Objective: Accurate surrogate modeling of knee joint contact mechanics is important for reconstructing stress distributions and identifying risk-relevant regions, yet the relative suitability of different modeling paradigms under practically relevant input-limited conditions remains unclear. Methods: Nine male soccer players performed 90\u00b0 change-of-direction trials. Finite element simulations driven by subject-specific joint posture and reaction forces were converted into graph-structured samples. Five surrogate architectures representing local diffusion, history-context enhancement, hierarchical multi-scale modeling, explicit global interaction, and local-global hybridization were compared using three-fold cross-subject validation under full, pose-corrupted, load-corrupted, and minimal-input conditions. Performance was evaluated using full-field error, high-stress error, high-risk region overlap, and hotspot localization metrics. Results: The hybrid model achieved the best overall performance under full inputs and remained the most robust under pose- and load-corrupted conditions. Under minimal inputs, no single model dominated all metrics: the history-context model yielded lower overall and high-stress errors, the hybrid model better preserved high-risk region reconstruction, and the hierarchical model showed an advantage in hotspot localization. Conclusion: Evaluation of surrogate models for knee joint contact mechanics should shift from accuracy comparisons under ideal inputs to a comprehensive assessment of the preservation of risk-relevant information under realistic input constraints. Although the local-global hybrid model showed the best overall robustness, the optimal model under minimal-input conditions remained task-dependent." }, "2511.09457.pdf": { "id": "2511.09457v1", "title": "The trade-off between model flexibility and accuracy of the Expected Threat model in football", "year": "2025", "authors": [ "Koen W. van Arem", "Jakob S\u00f6hl", "Mirjam Bruinsma", "Geurt Jongbloed" ], "categories": [ "stat.AP", "stat.ME" ], "pdf_url": "https://arxiv.org/pdf/2511.09457v1", "abstract": "With an average football (soccer) match recording over 3,000 on-ball events, effective use of this event data is essential for practitioners at football clubs to obtain meaningful insights. Models can extract more information from this data, and explainable methods can make them more accessible to practitioners. The Expected Threat model has been praised for its explainability and offers an accessible option. However, selecting the grid size is a challenging key design choice that has to be made when applying the Expected Threat model. Using a finer grid leads to a more flexible model that can better distinguish between different situations, but the accuracy of the estimates deteriorates with a more flexible model. Consequently, practitioners face challenges in balancing the trade-off between model flexibility and model accuracy. In this study, the Expected Threat model is analyzed from a theoretical perspective and simulations are performed based on the Markov chain of the model to examine its behavior in practice. Our theoretical results establish an upper bound on the error of the Expected Threat model for different flexibilities. Based on the simulations, a more accurate characterization of the model's error is provided, improving over the theoretical bound. Finally, these insights are converted into a practical rule of thumb to help practitioners choose the right balance between the model flexibility and the desired accuracy of the Expected Threat model." }, "2508.09238.pdf": { "id": "2508.09238v1", "title": "ELASTIC: Event-Tracking Data Synchronization in Soccer Without Annotated Event Locations", "year": "2025", "authors": [ "Hyunsung Kim", "Hoyoung Choi", "Sangwoo Seo", "Tom Boomstra", "Jinsung Yoon", "Chanyoung Park" ], "categories": [ "cs.DB" ], "pdf_url": "https://arxiv.org/pdf/2508.09238v1", "abstract": "The integration of event and tracking data has become essential for advanced analysis in soccer. However, synchronizing these two modalities remains a significant challenge due to temporal and spatial inaccuracies in manually recorded event timestamps. Existing synchronizers typically rely on annotated event locations, which themselves are prone to spatial errors and thus can distort synchronization results. To address this issue, we propose ELASTIC (Event-Location-AgnoSTIC synchronizer), a synchronization framework that only uses features derived from tracking data. ELASTIC also explicitly detects the end times of pass-like events and separates the detection of major and minor events, which improves the completeness of the synchronized output and reduces error cascade across events. We annotated the ground truth timestamps of 2,134 events from three Eredivisie matches to measure the synchronization accuracy, and the experimental results demonstrate that ELASTIC outperforms existing synchronizers by a large margin." }, "2504.10106.pdf": { "id": "2504.10106v1", "title": "SoccerNet-v3D: Leveraging Sports Broadcast Replays for 3D Scene Understanding", "year": "2025", "authors": [ "Marc Guti\u00e9rrez-P\u00e9rez", "Antonio Agudo" ], "categories": [ "cs.CV", "cs.AI" ], "pdf_url": "https://arxiv.org/pdf/2504.10106v1", "abstract": "Sports video analysis is a key domain in computer vision, enabling detailed spatial understanding through multi-view correspondences. In this work, we introduce SoccerNet-v3D and ISSIA-3D, two enhanced and scalable datasets designed for 3D scene understanding in soccer broadcast analysis. These datasets extend SoccerNet-v3 and ISSIA by incorporating field-line-based camera calibration and multi-view synchronization, enabling 3D object localization through triangulation. We propose a monocular 3D ball localization task built upon the triangulation of ground-truth 2D ball annotations, along with several calibration and reprojection metrics to assess annotation quality on demand. Additionally, we present a single-image 3D ball localization method as a baseline, leveraging camera calibration and ball size priors to estimate the ball's position from a monocular viewpoint. To further refine 2D annotations, we introduce a bounding box optimization technique that ensures alignment with the 3D scene representation. Our proposed datasets establish new benchmarks for 3D soccer scene understanding, enhancing both spatial and temporal analysis in sports analytics. Finally, we provide code to facilitate access to our annotations and the generation pipelines for the datasets." }, "2407.00187.pdf": { "id": "2407.00187v1", "title": "SMPLOlympics: Sports Environments for Physically Simulated Humanoids", "year": "2024", "authors": [ "Zhengyi Luo", "Jiashun Wang", "Kangni Liu", "Haotian Zhang", "Chen Tessler", "Jingbo Wang", "Ye Yuan", "Jinkun Cao", "Zihui Lin", "Fengyi Wang", "Jessica Hodgins", "Kris Kitani" ], "categories": [ "cs.RO", "cs.CV", "cs.GR" ], "pdf_url": "https://arxiv.org/pdf/2407.00187v1", "abstract": "We present SMPLOlympics, a collection of physically simulated environments that allow humanoids to compete in a variety of Olympic sports. Sports simulation offers a rich and standardized testing ground for evaluating and improving the capabilities of learning algorithms due to the diversity and physically demanding nature of athletic activities. As humans have been competing in these sports for many years, there is also a plethora of existing knowledge on the preferred strategy to achieve better performance. To leverage these existing human demonstrations from videos and motion capture, we design our humanoid to be compatible with the widely-used SMPL and SMPL-X human models from the vision and graphics community. We provide a suite of individual sports environments, including golf, javelin throw, high jump, long jump, and hurdling, as well as competitive sports, including both 1v1 and 2v2 games such as table tennis, tennis, fencing, boxing, soccer, and basketball. Our analysis shows that combining strong motion priors with simple rewards can result in human-like behavior in various sports. By providing a unified sports benchmark and baseline implementation of state and reward designs, we hope that SMPLOlympics can help the control and animation communities achieve human-like and performant behaviors." }, "2404.06626.pdf": { "id": "2404.06626v1", "title": "Why is soccer so popular: Understanding underdog achievement and randomness in team ball sports", "year": "2024", "authors": [ "Luis Nunes Vicente", "Thaksheel Alleck", "Tommaso Giovannelli", "Roman Mitchell", "Ori Remen" ], "categories": [ "stat.AP" ], "pdf_url": "https://arxiv.org/pdf/2404.06626v1", "abstract": "In this paper, we examine team ball sports to investigate how the likelihood of weaker teams winning against stronger ones, referred to as underdog achievement, is influenced by inherent randomness factors that affect match outcomes in such sports. To address our research question, we collected data on match scores and computed corresponding team rankings from major international competitions for 12 popular team ball sports: basketball, cricket, field hockey, futsal, handball, ice hockey, lacrosse, roller hockey, rugby, soccer, volleyball, and water polo. Then, we developed an underdog achievement score to identify the sports with the highest occurrences of weaker teams prevailing over stronger ones, and we designed a randomness model consisting of factors that contribute to unexpected match outcomes within each sport. Our findings indicate that soccer is among the sports in which a weaker team is most likely to win. Through principal component analysis (PCA) and correlation analysis, we demonstrate that our randomness model can explain such a phenomenon, showing that the underdog achievement can be attributed to numerous factors that can randomly influence match outcomes." }, "2401.09942.pdf": { "id": "2401.09942v1", "title": "Multi-task Learning for Joint Re-identification, Team Affiliation, and Role Classification for Sports Visual Tracking", "year": "2024", "authors": [ "Amir M. Mansourian", "Vladimir Somers", "Christophe De Vleeschouwer", "Shohreh Kasaei" ], "categories": [ "cs.CV", "cs.AI" ], "pdf_url": "https://arxiv.org/pdf/2401.09942v1", "abstract": "Effective tracking and re-identification of players is essential for analyzing soccer videos. But, it is a challenging task due to the non-linear motion of players, the similarity in appearance of players from the same team, and frequent occlusions. Therefore, the ability to extract meaningful embeddings to represent players is crucial in developing an effective tracking and re-identification system. In this paper, a multi-purpose part-based person representation method, called PRTreID, is proposed that performs three tasks of role classification, team affiliation, and re-identification, simultaneously. In contrast to available literature, a single network is trained with multi-task supervision to solve all three tasks, jointly. The proposed joint method is computationally efficient due to the shared backbone. Also, the multi-task learning leads to richer and more discriminative representations, as demonstrated by both quantitative and qualitative results. To demonstrate the effectiveness of PRTreID, it is integrated with a state-of-the-art tracking method, using a part-based post-processing module to handle long-term tracking. The proposed tracking method outperforms all existing tracking methods on the challenging SoccerNet tracking dataset." }, "2311.10361.pdf": { "id": "2311.10361v2", "title": "Video-based Sequential Bayesian Homography Estimation for Soccer Field Registration", "year": "2023", "authors": [ "Paul J. Claasen", "J. P. de Villiers" ], "categories": [ "cs.CV" ], "pdf_url": "https://arxiv.org/pdf/2311.10361v2", "abstract": "A novel Bayesian framework is proposed, which explicitly relates the homography of one video frame to the next through an affine transformation while explicitly modelling keypoint uncertainty. The literature has previously used differential homography between subsequent frames, but not in a Bayesian setting. In cases where Bayesian methods have been applied, camera motion is not adequately modelled, and keypoints are treated as deterministic. The proposed method, Bayesian Homography Inference from Tracked Keypoints (BHITK), employs a two-stage Kalman filter and significantly improves existing methods. Existing keypoint detection methods may be easily augmented with BHITK. It enables less sophisticated and less computationally expensive methods to outperform the state-of-the-art approaches in most homography evaluation metrics. Furthermore, the homography annotations of the WorldCup and TS-WorldCup datasets have been refined using a custom homography annotation tool that has been released for public use. The refined datasets are consolidated and released as the consolidated and refined WorldCup (CARWC) dataset." }, "2311.05025.pdf": { "id": "2311.05025v4", "title": "Unbiased Kinetic Langevin Monte Carlo with Inexact Gradients", "year": "2023", "authors": [ "Neil K. Chada", "Benedict Leimkuhler", "Daniel Paulin", "Peter A. Whalley" ], "categories": [ "stat.CO", "math.NA", "stat.ME", "stat.ML" ], "pdf_url": "https://arxiv.org/pdf/2311.05025v4", "abstract": "We present an unbiased method for Bayesian posterior means based on kinetic Langevin dynamics that combines advanced splitting methods with enhanced gradient approximations. Our approach avoids Metropolis correction by coupling Markov chains at different discretization levels in a multilevel Monte Carlo approach. Theoretical analysis demonstrates that our proposed estimator is unbiased, attains finite variance, and satisfies a central limit theorem. It can achieve accuracy $\u03b5>0$ for estimating expectations of Lipschitz functions in $d$ dimensions with $\\mathcal{O}(d^{1/4}\u03b5^{-2})$ expected gradient evaluations, without assuming warm start. We exhibit similar bounds using both approximate and stochastic gradients, and our method's computational cost is shown to scale independently of the size of the dataset. The proposed method is tested using a multinomial regression problem on the MNIST dataset and a Poisson regression model for soccer scores. Experiments indicate that the number of gradient evaluations per effective sample is independent of dimension, even when using inexact gradients. For product distributions, we give dimension-independent variance bounds. Our results demonstrate that in large-scale applications, the unbiased algorithm we present can be 2-3 orders of magnitude more efficient than the ``gold-standard\" randomized Hamiltonian Monte Carlo." }, "2311.03745.pdf": { "id": "2311.03745v2", "title": "Unsupervised Video Summarization via Iterative Training and Simplified GAN", "year": "2023", "authors": [ "Hanqing Li", "Diego Klabjan", "Jean Utke" ], "categories": [ "cs.CV", "cs.LG" ], "pdf_url": "https://arxiv.org/pdf/2311.03745v2", "abstract": "This paper introduces a new, unsupervised method for automatic video summarization using ideas from generative adversarial networks but eliminating the discriminator, having a simple loss function, and separating training of different parts of the model. An iterative training strategy is also applied by alternately training the reconstructor and the frame selector for multiple iterations. Furthermore, a trainable mask vector is added to the model in summary generation during training and evaluation. The method also includes an unsupervised model selection algorithm. Results from experiments on two public datasets (SumMe and TVSum) and four datasets we created (Soccer, LoL, MLB, and ShortMLB) demonstrate the effectiveness of each component on the model performance, particularly the iterative training strategy. Evaluations and comparisons with the state-of-the-art methods highlight the advantages of the proposed method in performance, stability, and training efficiency." }, "2310.14469.pdf": { "id": "2310.14469v1", "title": "Player Re-Identification Using Body Part Appearences", "year": "2023", "authors": [ "Mahesh Bhosale", "Abhishek Kumar", "David Doermann" ], "categories": [ "cs.CV" ], "pdf_url": "https://arxiv.org/pdf/2310.14469v1", "abstract": "We propose a neural network architecture that learns body part appearances for soccer player re-identification. Our model consists of a two-stream network (one stream for appearance map extraction and the other for body part map extraction) and a bilinear-pooling layer that generates and spatially pools the body part map. Each local feature of the body part map is obtained by a bilinear mapping of the corresponding local appearance and body part descriptors. Our novel representation yields a robust image-matching feature map, which results from combining the local similarities of the relevant body parts with the weighted appearance similarity. Our model does not require any part annotation on the SoccerNet-V3 re-identification dataset to train the network. Instead, we use a sub-network of an existing pose estimation network (OpenPose) to initialize the part substream and then train the entire network to minimize the triplet loss. The appearance stream is pre-trained on the ImageNet dataset, and the part stream is trained from scratch for the SoccerNet-V3 dataset. We demonstrate the validity of our model by showing that it outperforms state-of-the-art models such as OsNet and InceptionNet." }, "2303.15931.pdf": { "id": "2303.15931v1", "title": "FC Portugal 3D Simulation Team: Team Description Paper 2020", "year": "2023", "authors": [ "Nuno Lau", "Luis Paulo Reis", "David Simoes", "Mohammadreza Kasaei. Miguel Abreu", "Tiago Silva", "Francisco Resende" ], "categories": [ "cs.RO" ], "pdf_url": "https://arxiv.org/pdf/2303.15931v1", "abstract": "The FC Portugal 3D team is developed upon the structure of our previous Simulation league 2D/3D teams and our standard platform league team. Our research concerning the robot low-level skills is focused on developing behaviors that may be applied on real robots with minimal adaptation using model-based approaches. Our research on high-level soccer coordination methodologies and team playing is mainly focused on the adaptation of previously developed methodologies from our 2D soccer teams to the 3D humanoid environment and on creating new coordination methodologies based on the previously developed ones. The research-oriented development of our team has been pushing it to be one of the most competitive over the years (World champion in 2000 and Coach Champion in 2002, European champion in 2000 and 2001, Coach 2nd place in 2003 and 2004, European champion in Rescue Simulation and Simulation 3D in 2006, World Champion in Simulation 3D in Bremen 2006 and European champion in 2007, 2012, 2013, 2014 and 2015). This paper describes some of the main innovations of our 3D simulation league team during the last years. A new generic framework for reinforcement learning tasks has also been developed. The current research is focused on improving the above-mentioned framework by developing new learning algorithms to optimize low-level skills, such as running and sprinting. We are also trying to increase student contact by providing reinforcement learning assignments to be completed using our new framework, which exposes a simple interface without sharing low-level implementation details." }, "2211.12334.pdf": { "id": "2211.12334v1", "title": "A Graph-Based Method for Soccer Action Spotting Using Unsupervised Player Classification", "year": "2022", "authors": [ "Alejandro Cartas", "Coloma Ballester", "Gloria Haro" ], "categories": [ "cs.CV", "cs.AI", "cs.MM" ], "pdf_url": "https://arxiv.org/pdf/2211.12334v1", "abstract": "Action spotting in soccer videos is the task of identifying the specific time when a certain key action of the game occurs. Lately, it has received a large amount of attention and powerful methods have been introduced. Action spotting involves understanding the dynamics of the game, the complexity of events, and the variation of video sequences. Most approaches have focused on the latter, given that their models exploit the global visual features of the sequences. In this work, we focus on the former by (a) identifying and representing the players, referees, and goalkeepers as nodes in a graph, and by (b) modeling their temporal interactions as sequences of graphs. For the player identification, or player classification task, we obtain an accuracy of 97.72% in our annotated benchmark. For the action spotting task, our method obtains an overall performance of 57.83% average-mAP by combining it with other audiovisual modalities. This performance surpasses similar graph-based methods and has competitive results with heavy computing methods. Code and data are available at https://github.com/IPCV/soccer_action_spotting." }, "2210.16474.pdf": { "id": "2210.16474v2", "title": "Controlling ball progression in soccer", "year": "2022", "authors": [ "Catherine Pfaff", "Emily Hunter", "Haozhi Hong", "Daniel Forestell", "Ari Fialkov", "Zoey Drassinower", "Timothy Chan" ], "categories": [ "physics.soc-ph" ], "pdf_url": "https://arxiv.org/pdf/2210.16474v2", "abstract": "In this paper, we examine how soccer players can use their spatial relationships to control parts of the field and safely move play up the field via chains of ``safe configurations,'' i.e. configurations of players on a team ensuring the possessor of the ball has a collection of open passing options all connected by open passing lanes. An underlying philosophy behind our work is that it is most difficult to disrupt an attacking team's progression forward (with the ball) when this attacking team has multiple ``good'' options of how to proceed at each moment in time. We provide some evidence of this. Our main construction is a directed weighted graph where the nodes encode the configurations of players, the directed edges encode transformations between these configurations, and the weights encode the relative frequencies of the transformations. We conclude with a few applications and proposed further investigations. We believe that our work can serve as a launching platform for significant further investigation into how teams can ``safely progress'' the ball up the field, strategy development, and sophisticated decision making metrics. For coaches and players, we aim to streamline the process of moving safely up the field. In particular, we aim to construct a framework for creating new successful patterns of movement and aim that our framework allows for the development of new strategy. At the same time, our work could help identify configurations on the field which often result in turnovers or that allow for a lot of strategic flexibility (also impeding defensive containment of the attacking team). *Because the contributions of the female authors to this paper were by no means less than those of the male authors, we have chosen to reverse convention and to list the authors in reverse alphabetical order to ensure that the female authors were not listed only after the male authors." }, "2209.12248.pdf": { "id": "2209.12248v1", "title": "D$^{\\bf{3}}$: Duplicate Detection Decontaminator for Multi-Athlete Tracking in Sports Videos", "year": "2022", "authors": [ "Rui He", "Zehua Fu", "Qingjie Liu", "Yunhong Wang", "Xunxun Chen" ], "categories": [ "cs.CV" ], "pdf_url": "https://arxiv.org/pdf/2209.12248v1", "abstract": "Tracking multiple athletes in sports videos is a very challenging Multi-Object Tracking (MOT) task, since athletes often have the same appearance and are intimately covered with each other, making a common occlusion problem becomes an abhorrent duplicate detection. In this paper, the duplicate detection is newly and precisely defined as occlusion misreporting on the same athlete by multiple detection boxes in one frame. To address this problem, we meticulously design a novel transformer-based Duplicate Detection Decontaminator (D$^3$) for training, and a specific algorithm Rally-Hungarian (RH) for matching. Once duplicate detection occurs, D$^3$ immediately modifies the procedure by generating enhanced boxes losses. RH, triggered by the team sports substitution rules, is exceedingly suitable for sports videos. Moreover, to complement the tracking dataset that without shot changes, we release a new dataset based on sports video named RallyTrack. Extensive experiments on RallyTrack show that combining D$^3$ and RH can dramatically improve the tracking performance with 9.2 in MOTA and 4.5 in HOTA. Meanwhile, experiments on MOT-series and DanceTrack discover that D$^3$ can accelerate convergence during training, especially save up to 80 percent of the original training time on MOT17. Finally, our model, which is trained only with volleyball videos, can be applied directly to basketball and soccer videos for MAT, which shows priority of our method. Our dataset is available at https://github.com/heruihr/rallytrack." }, "2205.07193.pdf": { "id": "2205.07193v2", "title": "How Much Does Home Field Advantage Matter in Soccer Games? A Causal Inference Approach for English Premier League Analysis", "year": "2022", "authors": [ "Chen Wang", "Katherine Price", "Hengrui Cai", "Weining Shen", "Zhanrui Cai", "Guanyu Hu" ], "categories": [ "stat.AP", "stat.ME" ], "pdf_url": "https://arxiv.org/pdf/2205.07193v2", "abstract": "In many sports, it is commonly believed that the home team has an advantage over the visiting team, known as the home field advantage. Yet its causal effect on team performance is largely unknown. In this paper, we propose a novel causal inference approach to study the causal effect of home field advantage in English Premier League. We develop a hierarchical causal model and show that both league level and team level causal effects are identifiable and can be conveniently estimated. We further develop an inference procedure for the proposed estimators and demonstrate its excellent numerical performance via simulation studies. We implement our method on the 2020-21 English Premier League data and assess the causal effect of home advantage on eleven summary statistics that measure the offensive and defensive performance and referee-assessed disciplinary outcomes. We find that the home field advantage resides more heavily in offensive statistics than it does in defensive or referee statistics. We also find evidence that teams that had lower rankings retain a higher home field advantage." }, "2202.12259.pdf": { "id": "2202.12259v1", "title": "Learning from the Pros: Extracting Professional Goalkeeper Technique from Broadcast Footage", "year": "2022", "authors": [ "Matthew Wear", "Ryan Beal", "Tim Matthews", "Tim Norman", "Sarvapali Ramchurn" ], "categories": [ "cs.CV", "cs.AI" ], "pdf_url": "https://arxiv.org/pdf/2202.12259v1", "abstract": "As an amateur goalkeeper playing grassroots soccer, who better to learn from than top professional goalkeepers? In this paper, we harness computer vision and machine learning models to appraise the save technique of professionals in a way those at lower levels can learn from. We train an unsupervised machine learning model using 3D body pose data extracted from broadcast footage to learn professional goalkeeper technique. Then, an \"expected saves\" model is developed, from which we can identify the optimal goalkeeper technique in different match contexts." }, "2111.04525.pdf": { "id": "2111.04525v1", "title": "D-Flow: A Real Time Spatial Temporal Model for Target Area Segmentation", "year": "2021", "authors": [ "Wentao Lu", "Claude Sammut" ], "categories": [ "cs.RO" ], "pdf_url": "https://arxiv.org/pdf/2111.04525v1", "abstract": "Semantic segmentation has attracted a large amount of attention in recent years. In robotics, segmentation can be used to identify a region of interest, or \\emph{target area}. For example, in the RoboCup Standard Platform League (SPL), segmentation separates the soccer field from the background and from players on the field. For satellite or vehicle applications, it is often necessary to find certain regions such as roads, bodies of water or kinds of terrain. In this paper, we propose a novel approach to real-time target area segmentation based on a newly designed spatial temporal network. The method operates under domain constraints defined by both the robot's hardware and its operating environment . The proposed network is able to run in real-time, working within the constraints of limited run time and computing power. This work is compared against other real time segmentation methods on a dataset generated by a Nao V6 humanoid robot simulating the RoboCup SPL competition. In this case, the target area is defined as the artificial grass field. The method is also tested on a maritime dataset collected by a moving vessel, where the aim is to separate the ocean region from the rest of the image. This dataset demonstrates that the proposed model can generalise to a variety of vision problems." }, "2108.00845.pdf": { "id": "2108.00845v1", "title": "Scoring a Goal optimally in a Soccer game under Liouville-like quantum gravity action", "year": "2021", "authors": [ "Paramahansa Pramanik", "Alan M. Polansky" ], "categories": [ "math.OC" ], "pdf_url": "https://arxiv.org/pdf/2108.00845v1", "abstract": "In this paper we present a new theoretical model to find out an optimal weight associated with a soccer player under the presence of a stochastic goal dynamics by using Feynman path integral method, where the action of a player is on $\\sqrt{8/3}$-Liouville Quantum Gravity surface. Before determine the optimal weight we first establish an Infinitary logic which can deal with infinite variables on the strategy space then, a quantum formula of this logic has been developed and finally, based on this we are able show the existence of a Lefschetz-Hopf fixed point of this game. As in a competitive tournament, all possible standard strategies to score goals are known to the opposition team, a player's action is stochastic in nature and they would have some comparative advantage to score goals. Furthermore, conditions like uncertainties due to rain, dribbling and passing skill of a player, whether the match is a day match or a day-night match, advantages of having home crowd and asymmetric information of action profiles have been considered on the way to determine the optimal weight." }, "2104.09333.pdf": { "id": "2104.09333v1", "title": "Camera Calibration and Player Localization in SoccerNet-v2 and Investigation of their Representations for Action Spotting", "year": "2021", "authors": [ "Anthony Cioppa", "Adrien Deli\u00e8ge", "Floriane Magera", "Silvio Giancola", "Olivier Barnich", "Bernard Ghanem", "Marc Van Droogenbroeck" ], "categories": [ "cs.CV" ], "pdf_url": "https://arxiv.org/pdf/2104.09333v1", "abstract": "Soccer broadcast video understanding has been drawing a lot of attention in recent years within data scientists and industrial companies. This is mainly due to the lucrative potential unlocked by effective deep learning techniques developed in the field of computer vision. In this work, we focus on the topic of camera calibration and on its current limitations for the scientific community. More precisely, we tackle the absence of a large-scale calibration dataset and of a public calibration network trained on such a dataset. Specifically, we distill a powerful commercial calibration tool in a recent neural network architecture on the large-scale SoccerNet dataset, composed of untrimmed broadcast videos of 500 soccer games. We further release our distilled network, and leverage it to provide 3 ways of representing the calibration results along with player localization. Finally, we exploit those representations within the current best architecture for the action spotting task of SoccerNet-v2, and achieve new state-of-the-art performances." }, "2102.11182.pdf": { "id": "2102.11182v4", "title": "The Soccer Game, bit by bit: An information-theoretic analysis", "year": "2021", "authors": [ "Luis Ramada Pereira", "Rui J. Lopes", "Jorge Lou\u00e7\u00e3", "Duarte Ara\u00fajo", "Jo\u00e3o Ramos" ], "categories": [ "cs.SI", "physics.soc-ph" ], "pdf_url": "https://arxiv.org/pdf/2102.11182v4", "abstract": "We modeled the dynamics of a soccer match based on a network representation where players are nodes discretely clustered into homogeneous groups. Players were grouped by physical proximity, supported by the intuitive notion that competing and same-team players use relative position as a key tactical tool to contribute to the team's objectives. The model was applied to a set of matches from a major European national football league, with players' coordinates sampled at 10Hz, resulting in approx. 60,000 network samples per match. We took an information theoretic approach to measuring distance between samples and used it as a proxy for the game dynamics. Significant correlations were found between measurements and key match events that are empirically known to result in players jostling for position, such as when striving to get unmarked or to mark. These events increase the information distance, while breaks in game play have the opposite effect. By analyzing the frequency spectrum of players' cluster transitions and their corresponding information distance, it is possible to build a comprehensive view of player's interactions, useful for training and strategy development. This analysis can be drilled down to the level of individual players by quantifying their contribution to cluster breakup and emergence, building an overall multi-level map that provides insights into the game dynamics, from the individual player, to the clusters of interacting players, all the way to the teams and their matches." }, "2008.13183.pdf": { "id": "2008.13183v4", "title": "A New Approach to Generalised Uncertainty Relations", "year": "2020", "authors": [ "Matthew J. Lake" ], "categories": [ "gr-qc", "hep-ph", "quant-ph" ], "pdf_url": "https://arxiv.org/pdf/2008.13183v4", "abstract": "We outline a new model in which generalised uncertainty relations are obtained without modified commutation relations. While existing models introduce modified phase space volumes for the canonical degrees of freedom, we introduce new degrees of freedom for the background geometry. The phase space is therefore enlarged but remains Euclidean. The spatial background is treated as a genuinely quantum object, with an associated state vector, and the model naturally gives rise to the extended generalised uncertainty principle (EGUP). Importantly, this approach solves (or rather, evades) well known problems associated with modified commutators, including violation of the equivalence principle, the `soccer ball' problem for multiparticle states, and the velocity dependence of the minimum length. However, it implies two radical conclusions. The first is that space must be quantised on a different scale to matter and the second is that the fundamental quanta of geometry are fermions. We explain how, in the context of the model, these do not contradict established results including the no go theorems for multiple quantisation constants, which still hold for species of material particles, and the spin-$2$ nature of gravitons." }, "2006.08189.pdf": { "id": "2006.08189v1", "title": "Estimation of Skill Distributions", "year": "2020", "authors": [ "Ali Jadbabaie", "Anuran Makur", "Devavrat Shah" ], "categories": [ "stat.ML", "cs.LG", "math.ST", "stat.ME" ], "pdf_url": "https://arxiv.org/pdf/2006.08189v1", "abstract": "In this paper, we study the problem of learning the skill distribution of a population of agents from observations of pairwise games in a tournament. These games are played among randomly drawn agents from the population. The agents in our model can be individuals, sports teams, or Wall Street fund managers. Formally, we postulate that the likelihoods of game outcomes are governed by the Bradley-Terry-Luce (or multinomial logit) model, where the probability of an agent beating another is the ratio between its skill level and the pairwise sum of skill levels, and the skill parameters are drawn from an unknown skill density of interest. The problem is, in essence, to learn a distribution from noisy, quantized observations. We propose a simple and tractable algorithm that learns the skill density with near-optimal minimax mean squared error scaling as $n^{-1+\\varepsilon}$, for any $\\varepsilon>0$, when the density is smooth. Our approach brings together prior work on learning skill parameters from pairwise comparisons with kernel density estimation from non-parametric statistics. Furthermore, we prove minimax lower bounds which establish minimax optimality of the skill parameter estimation technique used in our algorithm. These bounds utilize a continuum version of Fano's method along with a covering argument. We apply our algorithm to various soccer leagues and world cups, cricket world cups, and mutual funds. We find that the entropy of a learnt distribution provides a quantitative measure of skill, which provides rigorous explanations for popular beliefs about perceived qualities of sporting events, e.g., soccer league rankings. Finally, we apply our method to assess the skill distributions of mutual funds. Our results shed light on the abundance of low quality funds prior to the Great Recession of 2008, and the domination of the industry by more skilled funds after the financial crisis." }, "1905.04762.pdf": { "id": "1905.04762v1", "title": "Real-Time Kinodynamic Motion Planning for Omnidirectional Mobile Robot Soccer using Rapidly-Exploring Random Tree in Dynamic Environment with Moving Obstacles", "year": "2019", "authors": [ "Fahri Ali Rahman", "Igi Ardiyanto", "Adha Imam Cahyadi" ], "categories": [ "cs.RO" ], "pdf_url": "https://arxiv.org/pdf/1905.04762v1", "abstract": "RoboCup Middle Size League (RoboCup MSL) provides a standardized testbed for research on mobile robot navigation, multi-robot cooperation, communication and integration via robot soccer competition in which the environment is highly dynamic and adversarial. One of important research topic in such area is kinodynamic motion planning that plan the trajectory of the robot while avoiding obstacles and obeying its dynamics. Kinodynamic motion planning for omnidirectional robot based on kinodynamic-RRT* method is presented in this work. Trajectory tracking control to execute the planned trajectory is also considered in this work. Robot motion planning in translational and rotational direction are decoupled. Then we implemented kinodynamic-RRT* with double integrator model to plan the translational trajectory. The rotational trajectory is generated using minimum-time trajectory generator satisfying velocity and acceleration constraints. The planned trajectory is then tracked using PI-Control. To address changing environment, we developed concurrent sofware module for motion planning and trajectory tracking. The resulting system were applied and tested using RoboCup simulation system based on Robot Operating System (ROS). The simulation results that the motion planning system are able to generate collision-free trajectory and the trajectory tracking system are able to follow the generated trajectory. It is also shown that in highly dynamic environment the online scheme are able to re-plan the trajectory." }, "1903.01373.pdf": { "id": "1903.01373v4", "title": "$\u03b1$-Rank: Multi-Agent Evaluation by Evolution", "year": "2019", "authors": [ "Shayegan Omidshafiei", "Christos Papadimitriou", "Georgios Piliouras", "Karl Tuyls", "Mark Rowland", "Jean-Baptiste Lespiau", "Wojciech M. Czarnecki", "Marc Lanctot", "Julien Perolat", "Remi Munos" ], "categories": [ "cs.MA", "cs.GT" ], "pdf_url": "https://arxiv.org/pdf/1903.01373v4", "abstract": "We introduce $\u03b1$-Rank, a principled evolutionary dynamics methodology for the evaluation and ranking of agents in large-scale multi-agent interactions, grounded in a novel dynamical game-theoretic solution concept called Markov-Conley chains (MCCs). The approach leverages continuous- and discrete-time evolutionary dynamical systems applied to empirical games, and scales tractably in the number of agents, the type of interactions, and the type of empirical games (symmetric and asymmetric). Current models are fundamentally limited in one or more of these dimensions and are not guaranteed to converge to the desired game-theoretic solution concept (typically the Nash equilibrium). $\u03b1$-Rank provides a ranking over the set of agents under evaluation and provides insights into their strengths, weaknesses, and long-term dynamics. This is a consequence of the links we establish to the MCC solution concept when the underlying evolutionary model's ranking-intensity parameter, $\u03b1$, is chosen to be large, which exactly forms the basis of $\u03b1$-Rank. In contrast to the Nash equilibrium, which is a static concept based on fixed points, MCCs are a dynamical solution concept based on the Markov chain formalism, Conley's Fundamental Theorem of Dynamical Systems, and the core ingredients of dynamical systems: fixed points, recurrent sets, periodic orbits, and limit cycles. $\u03b1$-Rank runs in polynomial time with respect to the total number of pure strategy profiles, whereas computing a Nash equilibrium for a general-sum game is known to be intractable. We introduce proofs that not only provide a unifying perspective of existing continuous- and discrete-time evolutionary evaluation models, but also reveal the formal underpinnings of the $\u03b1$-Rank methodology. We empirically validate the method in several domains including AlphaGo, AlphaZero, MuJoCo Soccer, and Poker." }, "1901.00270.pdf": { "id": "1901.00270v1", "title": "Learning Humanoid Robot Motions Through Deep Neural Networks", "year": "2019", "authors": [ "Luckeciano Carvalho Melo", "Marcos Ricardo Omena Albuquerque Maximo", "Adilson Marques da Cunha" ], "categories": [ "cs.AI" ], "pdf_url": "https://arxiv.org/pdf/1901.00270v1", "abstract": "Controlling a high degrees of freedom humanoid robot is acknowledged as one of the hardest problems in Robotics. Due to the lack of mathematical models, an approach frequently employed is to rely on human intuition to design keyframe movements by hand, usually aided by graphical tools. In this paper, we propose a learning framework based on neural networks in order to mimic humanoid robot movements. The developed technique does not make any assumption about the underlying implementation of the movement, therefore both keyframe and model-based motions may be learned. The framework was applied in the RoboCup 3D Soccer Simulation domain and promising results were obtained using the same network architecture for several motions, even when copying motions from another teams." }, "1901.00172.pdf": { "id": "1901.00172v3", "title": "Supervised Multiscale Dimension Reduction for Spatial Interaction Networks", "year": "2019", "authors": [ "Shaobo Han", "David B. Dunson" ], "categories": [ "cs.LG", "cs.SI", "stat.ML" ], "pdf_url": "https://arxiv.org/pdf/1901.00172v3", "abstract": "We introduce a multiscale supervised dimension reduction method for SPatial Interaction Network (SPIN) data, which consist of a collection of spatially coordinated interactions. This type of predictor arises when the sampling unit of data is composed of a collection of primitive variables, each of them being essentially unique, so that it becomes necessary to group the variables in order to simplify the representation and enhance interpretability. In this paper, we introduce an empirical Bayes approach called spinlets, which first constructs a partitioning tree to guide the reduction over multiple spatial granularities, and then refines the representation of predictors according to the relevance to the response. We consider an inverse Poisson regression model and propose a new multiscale generalized double Pareto prior, which is induced via a tree-structured parameter expansion scheme. Our approach is motivated by an application in soccer analytics, in which we obtain compact vectorial representations and readily interpretable visualizations of the complex network objects, supervised by the response of interest." }, "1812.01895.pdf": { "id": "1812.01895v1", "title": "Computational Graph Approach for Detection of Composite Human Activities", "year": "2018", "authors": [ "Niko Reunanen", "Ville K\u00f6n\u00f6nen", "Hermanni H\u00e4lv\u00e4", "Jani M\u00e4ntyj\u00e4rvi", "Arttu L\u00e4ms\u00e4", "Jussi Liikka" ], "categories": [ "cs.NE" ], "pdf_url": "https://arxiv.org/pdf/1812.01895v1", "abstract": "Existing work in human activity detection classifies physical activities using a single fixed-length subset of a sensor signal. However, temporally consecutive subsets of a sensor signal are not utilized. This is not optimal for classifying physical activities (composite activities) that are composed of a temporal series of simpler activities (atomic activities). A sport consists of physical activities combined in a fashion unique to that sport. The constituent physical activities and the sport are not fundamentally different. We propose a computational graph architecture for human activity detection based on the readings of a triaxial accelerometer. The resulting model learns 1) a representation of the atomic activities of a sport and 2) to classify physical activities as compositions of the atomic activities. The proposed model, alongside with a set of baseline models, was tested for a simultaneous classification of eight physical activities (walking, nordic walking, running, soccer, rowing, bicycling, exercise bicycling and lying down). The proposed model obtained an overall mean accuracy of 77.91% (population) and 95.28% (personalized). The corresponding accuracies of the best baseline model were 73.52% and 90.03%. However, without combining consecutive atomic activities, the corresponding accuracies of the proposed model were 71.52% and 91.22%. The results show that our proposed model is accurate, outperforms the baseline models and learns to combine simple activities into complex activities. Composite activities can be classified as combinations of atomic activities. Our proposed architecture is a basis for accurate models in human activity detection." }, "1810.06394.pdf": { "id": "1810.06394v1", "title": "Parametrized Deep Q-Networks Learning: Reinforcement Learning with Discrete-Continuous Hybrid Action Space", "year": "2018", "authors": [ "Jiechao Xiong", "Qing Wang", "Zhuoran Yang", "Peng Sun", "Lei Han", "Yang Zheng", "Haobo Fu", "Tong Zhang", "Ji Liu", "Han Liu" ], "categories": [ "cs.LG", "cs.AI", "stat.ML" ], "pdf_url": "https://arxiv.org/pdf/1810.06394v1", "abstract": "Most existing deep reinforcement learning (DRL) frameworks consider either discrete action space or continuous action space solely. Motivated by applications in computer games, we consider the scenario with discrete-continuous hybrid action space. To handle hybrid action space, previous works either approximate the hybrid space by discretization, or relax it into a continuous set. In this paper, we propose a parametrized deep Q-network (P- DQN) framework for the hybrid action space without approximation or relaxation. Our algorithm combines the spirits of both DQN (dealing with discrete action space) and DDPG (dealing with continuous action space) by seamlessly integrating them. Empirical results on a simulation example, scoring a goal in simulated RoboCup soccer and the solo mode in game King of Glory (KOG) validate the efficiency and effectiveness of our method." }, "1712.02224.pdf": { "id": "1712.02224v1", "title": "Human Perception of Performance", "year": "2017", "authors": [ "Luca Pappalardo", "Paolo Cintia", "Dino Pedreschi", "Fosca Giannotti", "Albert-Laszlo Barabasi" ], "categories": [ "physics.soc-ph", "cs.AI", "physics.data-an", "stat.AP" ], "pdf_url": "https://arxiv.org/pdf/1712.02224v1", "abstract": "Humans are routinely asked to evaluate the performance of other individuals, separating success from failure and affecting outcomes from science to education and sports. Yet, in many contexts, the metrics driving the human evaluation process remain unclear. Here we analyse a massive dataset capturing players' evaluations by human judges to explore human perception of performance in soccer, the world's most popular sport. We use machine learning to design an artificial judge which accurately reproduces human evaluation, allowing us to demonstrate how human observers are biased towards diverse contextual features. By investigating the structure of the artificial judge, we uncover the aspects of the players' behavior which attract the attention of human judges, demonstrating that human evaluation is based on a noticeability heuristic where only feature values far from the norm are considered to rate an individual's performance." }, "1403.6822.pdf": { "id": "1403.6822v1", "title": "Comparison of Multi-agent and Single-agent Inverse Learning on a Simulated Soccer Example", "year": "2014", "authors": [ "Xiaomin Lin", "Peter A. Beling", "Randy Cogill" ], "categories": [ "cs.LG", "cs.GT" ], "pdf_url": "https://arxiv.org/pdf/1403.6822v1", "abstract": "We compare the performance of Inverse Reinforcement Learning (IRL) with the relative new model of Multi-agent Inverse Reinforcement Learning (MIRL). Before comparing the methods, we extend a published Bayesian IRL approach that is only applicable to the case where the reward is only state dependent to a general one capable of tackling the case where the reward depends on both state and action. Comparison between IRL and MIRL is made in the context of an abstract soccer game, using both a game model in which the reward depends only on state and one in which it depends on both state and action. Results suggest that the IRL approach performs much worse than the MIRL approach. We speculate that the underperformance of IRL is because it fails to capture equilibrium information in the manner possible in MIRL." }, "1310.5781.pdf": { "id": "1310.5781v1", "title": "RANSAC: Identification of Higher-Order Geometric Features and Applications in Humanoid Robot Soccer", "year": "2013", "authors": [ "Madison Flannery", "Shannon Fenn", "David Budden" ], "categories": [ "cs.RO", "cs.AI", "cs.CV" ], "pdf_url": "https://arxiv.org/pdf/1310.5781v1", "abstract": "The ability for an autonomous agent to self-localise is directly proportional to the accuracy and precision with which it can perceive salient features within its local environment. The identification of such features by recognising geometric profile allows robustness against lighting variations, which is necessary in most industrial robotics applications. This paper details a framework by which the random sample consensus (RANSAC) algorithm, often applied to parameter fitting in linear models, can be extended to identify higher-order geometric features. Goalpost identification within humanoid robot soccer is investigated as an application, with the developed system yielding an order-of-magnitude improvement in classification performance relative to a traditional histogramming methodology." }, "1704.03847.pdf": { "id": "1704.03847v1", "title": "Semantic3D.net: A new Large-scale Point Cloud Classification Benchmark", "year": "2017", "authors": [ "Timo Hackel", "Nikolay Savinov", "Lubor Ladicky", "Jan D. Wegner", "Konrad Schindler", "Marc Pollefeys" ], "categories": [ "cs.CV", "cs.LG", "cs.NE", "cs.RO" ], "pdf_url": "https://arxiv.org/pdf/1704.03847v1", "abstract": "This paper presents a new 3D point cloud classification benchmark data set with over four billion manually labelled points, meant as input for data-hungry (deep) learning methods. We also discuss first submissions to the benchmark that use deep convolutional neural networks (CNNs) as a work horse, which already show remarkable performance improvements over state-of-the-art. CNNs have become the de-facto standard for many tasks in computer vision and machine learning like semantic segmentation or object detection in images, but have no yet led to a true breakthrough for 3D point cloud labelling tasks due to lack of training data. With the massive data set presented in this paper, we aim at closing this data gap to help unleash the full potential of deep learning methods for 3D labelling tasks. Our semantic3D.net data set consists of dense point clouds acquired with static terrestrial laser scanners. It contains 8 semantic classes and covers a wide range of urban outdoor scenes: churches, streets, railroad tracks, squares, villages, soccer fields and castles. We describe our labelling interface and show that our data set provides more dense and complete point clouds with much higher overall number of labelled points compared to those already available to the research community. We further provide baseline method descriptions and comparison between methods submitted to our online system. We hope semantic3D.net will pave the way for deep learning methods in 3D point cloud labelling to learn richer, more general 3D representations, and first submissions after only a few months indicate that this might indeed be the case." }, "1611.02266.pdf": { "id": "1611.02266v2", "title": "Gaussian Attention Model and Its Application to Knowledge Base Embedding and Question Answering", "year": "2016", "authors": [ "Liwen Zhang", "John Winn", "Ryota Tomioka" ], "categories": [ "stat.ML", "cs.AI", "cs.CL", "cs.LG" ], "pdf_url": "https://arxiv.org/pdf/1611.02266v2", "abstract": "We propose the Gaussian attention model for content-based neural memory access. With the proposed attention model, a neural network has the additional degree of freedom to control the focus of its attention from a laser sharp attention to a broad attention. It is applicable whenever we can assume that the distance in the latent space reflects some notion of semantics. We use the proposed attention model as a scoring function for the embedding of a knowledge base into a continuous vector space and then train a model that performs question answering about the entities in the knowledge base. The proposed attention model can handle both the propagation of uncertainty when following a series of relations and also the conjunction of conditions in a natural way. On a dataset of soccer players who participated in the FIFA World Cup 2014, we demonstrate that our model can handle both path queries and conjunctive queries well." }, "1511.04143.pdf": { "id": "1511.04143v5", "title": "Deep Reinforcement Learning in Parameterized Action Space", "year": "2015", "authors": [ "Matthew Hausknecht", "Peter Stone" ], "categories": [ "cs.AI", "cs.LG", "cs.MA", "cs.NE" ], "pdf_url": "https://arxiv.org/pdf/1511.04143v5", "abstract": "Recent work has shown that deep neural networks are capable of approximating both value functions and policies in reinforcement learning domains featuring continuous state and action spaces. However, to the best of our knowledge no previous work has succeeded at using deep neural networks in structured (parameterized) continuous action spaces. To fill this gap, this paper focuses on learning within the domain of simulated RoboCup soccer, which features a small set of discrete action types, each of which is parameterized with continuous variables. The best learned agent can score goals more reliably than the 2012 RoboCup champion agent. As such, this paper represents a successful extension of deep reinforcement learning to the class of parameterized action space MDPs." }, "2607.18009.pdf": { "id": "2607.18009v1", "title": "Bayesian Conway-Maxwell-Poisson model with spike-and slab priors for dispersed count data with application to football scores", "year": "2026", "authors": [ "Nick Zhang", "Riccardo Rastelli", "Nial Friel" ], "categories": [ "stat.ME", "stat.AP" ], "pdf_url": "https://arxiv.org/pdf/2607.18009v1", "abstract": "Statistical modeling for goals scored in football is typically achieved using the Poisson distribution and its variants. Here we propose a Bayesian framework for modeling under- and over-dispersion in count data by combining the Conway-Maxwell-Poisson (CMP) likelihood with a spikeand-slab (SAS) prior on unit-specific dispersion parameters. The proposed methodology generalizes Poisson-based count data models by treating equidispersion as an explicit baseline, and offering probabilistic quantification of departures from this regime, while simultaneously estimating their magnitude. Posterior inference is performed through a tailored Metropolis-within-Gibbs sampler that handles the doubly-intractable likelihood and provides efficient posterior exploration. The new method is examined using simulated data to confirm its ability to capture non-equidispersion, and applied to English Premier League (EPL) data. Dispersion is modeled at the team level and linked to goal-scoring behavior, and allows for thresholding mechanisms to distinguish teams based on their posterior probability of non-equidispersion. The results reveal heterogeneities in team-specific dispersion in the EPL, and demonstrate improvements in both model fit and predictive performance with respect to the standard Poisson model." }, "2606.21814.pdf": { "id": "2606.21814v1", "title": "Ranking football teams via the higher-order decomposition of performance networks", "year": "2026", "authors": [ "A. Chacoma", "J. I. Perotti", "O. V. Billoni" ], "categories": [ "physics.soc-ph", "physics.data-an" ], "pdf_url": "https://arxiv.org/pdf/2606.21814v1", "abstract": "We propose a unified methodological framework to quantify team performance in elite football by combining event-level performance metrics, higher-order network representations, and algebraic ranking methods. Using data from the 2017--2018 season of the five major European leagues, we construct metric-specific weighted graphs in which teams are connected through relative performance indicators. These graphs are analyzed via Hodge decomposition, and the gradient component is used to derive metric-based team ratings. The resulting rankings are systematically compared with the true league standings using Pearson and Kendall correlation measures, revealing strong metric- and league-dependent effects. Furthermore, by analyzing the ratio between solenoidal and total flow energies, we show that local cyclic dynamics structurally limit the gradient component's capacity to reconstruct the ranking. This topological inconsistency acts as a structural fingerprint of each league's ``competition style'' successfully mapping the studied systems into distinct regimes: highly hierarchical structures (England and Italy), tactical parity driven by generalized loops (Germany), and pockets of localized chaos (France and Spain). Lastly, we introduce a composite rating obtained as a parsimonious linear combination of metric-based ratings, optimized separately for each league. This composite approach significantly improves predictive power and allows the relative importance of different performance indicators to be quantified in a league-specific manner. Our results demonstrate how higher-order network methods provide a flexible and interpretable framework to uncover latent performance structures in football, offering a complementary perspective to outcome-based rankings and a general approach applicable to other oppositional sports." }, "2605.30209.pdf": { "id": "2605.30209v2", "title": "Integrity at Stake: Statistical Detection of Anomalous In-Play Betting in Football", "year": "2026", "authors": [ "David Winkelmann", "Maya Vienken", "Christian Deutscher", "Roland Langrock" ], "categories": [ "econ.GN", "stat.AP" ], "pdf_url": "https://arxiv.org/pdf/2605.30209v2", "abstract": "Match-fixing undermines the integrity of sport by eroding public trust and threatening the financial sustainability of clubs and leagues. The expansion of sports betting markets has created new incentives and opportunities for manipulation, calling for rigorous, data-driven monitoring tools. Football, which accounts for the largest share of global betting turnover, remains particularly exposed: integrity reports continue to flag several suspicious matches, with past scandals in Italy and Turkey underlining the problem's persistence. This study uses high-frequency in-play (live) betting data from the Italian Serie B (seasons 2018/19--2020/21) to explore statistical approaches for detecting unusual betting behaviour. A state-space modelling framework is employed to describe standard betting market dynamics and to predict expected betting volumes conditional on match characteristics. Deviations from these expectations can then be analysed using outlier detection techniques to identify potentially suspicious betting activities. The results demonstrate how statistical modelling can contribute to the early identification of unusual betting patterns, thereby supporting integrity assurance in in-play sports betting markets." }, "2510.17641.pdf": { "id": "2510.17641v6", "title": "Are penalty shootouts better than a coin toss? Evidence from international club football in Europe", "year": "2025", "authors": [ "L\u00e1szl\u00f3 Csat\u00f3", "D\u00f3ra Gr\u00e9ta Petr\u00f3czy" ], "categories": [ "econ.GN", "physics.soc-ph", "stat.AP" ], "pdf_url": "https://arxiv.org/pdf/2510.17641v6", "abstract": "Penalty shootouts play a crucial role in the knockout stage of major football tournaments. Their importance has substantially increased from the 2021/22 season, when the Union of European Football Associations (UEFA) scrapped the away goals rule. Our paper examines whether the outcome of a penalty shootout can be predicted in UEFA club competitions. Based on all shootouts between 2000 and 2025, no evidence is found for the effect of the kicking order, the field of the match, or psychological momentum. In contrast to previous results, we do not detect any relationship between shootout success and relative team strength, quantified by differences in Elo ratings and the implied winning probability. Thus, the hypothesis that penalty shootouts are close to a coin toss in international competitions for European football clubs cannot be rejected." }, "2502.00560.pdf": { "id": "2502.00560v4", "title": "Solving Football by Exploiting Equilibrium Structure of 2p0s Differential Games with One-Sided Information", "year": "2025", "authors": [ "Mukesh Ghimire", "Lei Zhang", "Zhe Xu", "Yi Ren" ], "categories": [ "cs.GT" ], "pdf_url": "https://arxiv.org/pdf/2502.00560v4", "abstract": "For a two-player imperfect-information extensive-form game (IIEFG) with $K$ time steps and a player action space of size $U$, the game tree complexity is $U^{2K}$, causing existing IIEFG solvers to struggle with large or infinite $(U,K)$, e.g., differential games with continuous action spaces. To partially address this scalability challenge, we focus on an important class of 2p0s games where the informed player (P1) knows the payoff while the uninformed player (P2) only has a belief over the set of $I$ possible payoffs. Such games encompass a wide range of scenarios in sports, defense, cybersecurity, and finance. We prove that under mild conditions, P1's (resp. P2's) equilibrium strategy at any infostate concentrates on at most $I$ (resp. $I+1$) action prototypes. When $I\\ll U$, this equilibrium structure causes the game tree complexity to collapse to $I^K$ for P1 when P2 plays best responses, and $(I+1)^K$ for P2 in a dual game where P1 plays best responses. We then show that exploiting this structure in model-free multiagent reinforcement learning and model predictive control leads to significant improvements in learning accuracy and efficiency from SOTA IIEFG solvers. Our demonstration solves a 22-player football game with continuous action spaces and $K=10$ time steps, where the offense team needs to strategically conceal their play until a critical moment in order to exploit information advantage. Code is available at https://github.com/ghimiremukesh/cams/tree/iclr" }, "2107.02029.pdf": { "id": "2107.02029v2", "title": "A physics-based algorithm to perform predictions in football leagues", "year": "2021", "authors": [ "Eduardo V. Stock", "Roberto da Silva", "Henrique A. Fernandes" ], "categories": [ "physics.soc-ph" ], "pdf_url": "https://arxiv.org/pdf/2107.02029v2", "abstract": "In this work, we extended a stochastic model for football leagues based on the team's potential [R. da Silva et al. Comput. Phys. Commun. \\textbf{184} 661--670 (2013)] for making predictions instead of only performing a successful characterization of the statistics on the punctuation of the real leagues. Our adaptation considers the advantage of playing at home when considering the potential of the home and away teams. The algorithm predicts the tournament's outcome by using the market value or/and the ongoing team's performance as initial conditions in the context of Monte Carlo simulations. We present and compare our results to the worldwide known SPI predictions performed by the \"FiveThirtyEight\" project. The results show that the algorithm can deliver good predictions even with a few ingredients and in more complicated seasons like the 2020 editions where the matches were played without fans in the stadiums." }, "2101.10597.pdf": { "id": "2101.10597v1", "title": "The Probabilistic Final Standing Calculator: a fair stochastic tool to handle abruptly stopped football seasons", "year": "2021", "authors": [ "Hans Van Eetvelde", "Lars Magnus Hvattum", "Christophe Ley" ], "categories": [ "stat.AP" ], "pdf_url": "https://arxiv.org/pdf/2101.10597v1", "abstract": "The COVID-19 pandemic has left its marks in the sports world, forcing the full-stop of all sports-related activities in the first half of 2020. Football leagues were suddenly stopped and each country was hesitating between a relaunch of the competition and a premature ending. Some opted for the latter option, and took as the final standing of the season the ranking from the moment the competition got interrupted. This decision has been perceived as unfair, especially by those teams who had remaining matches against easier opponents. In this paper, we introduce a tool to calculate in a fairer way the final standings of domestic leagues that have to stop prematurely: our Probabilistic Final Standing Calculator (PFSC). It is based on a stochastic model taking into account the results of the matches played and simulating the remaining matches, yielding the probabilities for the various possible final rankings. We have compared our PFSC with state-of-the-art prediction models, using previous seasons which we pretend to stop at different points in time. We illustrate our PFSC by showing how a probabilistic ranking of the French Ligue 1 in the stopped 2019-2020 season could have led to alternative, potentially fairer, decisions on the final standing." }, "1908.00823.pdf": { "id": "1908.00823v2", "title": "Generalised Joint Regression for Count Data with a Focus on Modelling Football Matches", "year": "2019", "authors": [ "Hendrik van der Wurp", "Andreas Groll", "Thomas Kneib", "Giampiero Marra", "Rosalba Radice" ], "categories": [ "stat.AP", "stat.ME" ], "pdf_url": "https://arxiv.org/pdf/1908.00823v2", "abstract": "We propose a versatile joint regression framework for count responses. The method is implemented in the R add-on package GJRM and allows for modelling linear and non-linear dependence through the use of several copulae. Moreover, the parameters of the marginal distributions of the count responses and of the copula can be specified as flexible functions of covariates. Motivated by a football application, we also discuss an extension which forces the regression coefficients of the marginal (linear) predictors to be equal via a suitable penalisation. Model fitting is based on a trust region algorithm which estimates simultaneously all the parameters of the joint models. We investigate the proposal's empirical performance in two simulation studies, the first one designed for arbitrary count data, the other one reflecting football-specific settings. Finally, the method is applied to FIFA World Cup data, showing its competitiveness to the standard approach with regard to predictive performance." }, "2004.08607.pdf": { "id": "2004.08607v1", "title": "Accumulator Bet Selection Through Stochastic Diffusion Search", "year": "2020", "authors": [ "Nassim Dehouche" ], "categories": [ "cs.AI", "math.OC" ], "pdf_url": "https://arxiv.org/pdf/2004.08607v1", "abstract": "An accumulator is a bet that presents a rather unique payout structure, in that it combines multiple bets into a wager that can generate a total payout given by the multiplication of the individual odds of its parts. These potentially important returns come however at an increased risk of a loss. Indeed, the presence of a single incorrect bet in this selection would make the whole accumulator lose. The complexity of selecting a set of matches to place an accumulator bet on, as well as the number of opportunities to identify winning combinations have both dramatically increased with the easier access to online and offline bookmakers that bettors have nowadays. We address this relatively under-studied combinatorial aspect of sports betting, and propose a binary optimization model for the problem of selecting the most promising combinations of matches, in terms of their total potential payout and probability of a win, to form an accumulator bet. The results of an ongoing computational experiment, in which our model is applied to real data pertaining to the four main football leagues in the world over a complete season, are presented and compared to those of single bet selection methods." }, "1403.6508.pdf": { "id": "1403.6508v3", "title": "Multi-agent Inverse Reinforcement Learning for Two-person Zero-sum Games", "year": "2014", "authors": [ "Xiaomin Lin", "Peter A. Beling", "Randy Cogill" ], "categories": [ "cs.GT", "cs.AI", "cs.LG" ], "pdf_url": "https://arxiv.org/pdf/1403.6508v3", "abstract": "The focus of this paper is a Bayesian framework for solving a class of problems termed multi-agent inverse reinforcement learning (MIRL). Compared to the well-known inverse reinforcement learning (IRL) problem, MIRL is formalized in the context of stochastic games, which generalize Markov decision processes to game theoretic scenarios. We establish a theoretical foundation for competitive two-agent zero-sum MIRL problems and propose a Bayesian solution approach in which the generative model is based on an assumption that the two agents follow a minimax bi-policy. Numerical results are presented comparing the Bayesian MIRL method with two existing methods in the context of an abstract soccer game. Investigation centers on relationships between the extent of prior information and the quality of learned rewards. Results suggest that covariance structure is more important than mean value in reward priors." }, "1201.6626.pdf": { "id": "1201.6626v1", "title": "Learning RoboCup-Keepaway with Kernels", "year": "2012", "authors": [ "Tobias Jung", "Daniel Polani" ], "categories": [ "cs.AI", "cs.LG", "cs.MA" ], "pdf_url": "https://arxiv.org/pdf/1201.6626v1", "abstract": "We apply kernel-based methods to solve the difficult reinforcement learning problem of 3vs2 keepaway in RoboCup simulated soccer. Key challenges in keepaway are the high-dimensionality of the state space (rendering conventional discretization-based function approximation like tilecoding infeasible), the stochasticity due to noise and multiple learning agents needing to cooperate (meaning that the exact dynamics of the environment are unknown) and real-time learning (meaning that an efficient online implementation is required). We employ the general framework of approximate policy iteration with least-squares-based policy evaluation. As underlying function approximator we consider the family of regularization networks with subset of regressors approximation. The core of our proposed solution is an efficient recursive implementation with automatic supervised selection of relevant basis functions. Simulation results indicate that the behavior learned through our approach clearly outperforms the best results obtained earlier with tilecoding by Stone et al. (2005)." }, "2610.26892.pdf": { "id": "2610.26892v1", "title": "Minimal-length kinematics of coherent light and a proposed two-color vacuum chronoscopy", "year": "2026", "authors": [ "Zhi-Wei Wang", "Saurya Das", "Samuel L. Braunstein" ], "categories": [ "gr-qc", "hep-ex", "hep-ph", "hep-th", "quant-ph" ], "pdf_url": "https://arxiv.org/pdf/2610.26892v1", "abstract": "The unification of quantum theory and general relativity suggests the existence of a minimal measurable length, leading to a generalized uncertainty principle (GUP) and an energy-dependent vacuum group velocity. Testing these effects with macroscopic objects typically encounters the ``soccer-ball problem,'' where applying Planck-scale deformations to the total momentum predicts observationally ruled-out velocity deviations. We adopt the QFT-natural prescription in which the GUP modifies single-mode commutators, and verify by explicit calculation that the macroscopic group velocity of a multimode coherent pulse reduces to the single-photon GUP velocity evaluated at the central wavelength, multiplied by the classical Gaussian-beam factor, with no dependence on the total pulse energy. The energy--momentum relation $E=pc$ itself is unmodified; the deformation enters through the velocity--momentum relation and is equivalent to a modified dispersion relation in the wave vector, with a quadratic coefficient that depends on which of the two variables is used. The soccer-ball problem $\\,$---$\\,$ which arises only when the GUP is applied to the center-of-mass momentum of a composite system $\\,$---$\\,$ is therefore absent for coherent light, and macroscopic petawatt pulses probe the same parameter space as single high-energy photons. Building on this framework, we propose a two-color vacuum chronoscopy experiment using a short-pulse petawatt laser and its high-harmonic generation. By systematically varying the macroscopic beam waist, our setup separates Planck-scale kinematic shifts from classical diffraction delays, establishing a controlled, source-independent terrestrial methodology whose sensitivity we assess against current astrophysical bounds and against the dominant waist-independent systematics that limit it, residual-gas dispersion and the harmonic birth-time offset." }, "2610.14042.pdf": { "id": "2610.14042v1", "title": "Reinforcement Learning from Real-Robot Interaction without High-Quality State Estimates", "year": "2026", "authors": [ "Ryan Gao", "Ashvin Anilkumar", "Nicholas Corrado", "William Cong", "Josiah Hanna" ], "categories": [ "cs.RO", "cs.AI" ], "pdf_url": "https://arxiv.org/pdf/2610.14042v1", "abstract": "A longstanding goal for robot reinforcement learning (RL) is that robots learn from their own physical experience, as humans and animals do. A key obstacle to learning directly from physical experience, especially outside instrumented lab environments, is that the state of the world is usually unobservable to the learner and must be estimated from on-board sensing. Errors in that estimate corrupt both the policy input and potentially the reward signal. We study this challenge using the contact-rich task of a bipedal robot learning to score soccer goals. We estimate all key state variables on-board with an unscented Kalman filter (UKF) and no external instrumentation (e.g., motion capture or overhead cameras). Because the UKF may incorrectly estimate state when the ball is unobserved, we learn a new state representation with a recurrent state-space model (RSSM) that, critically, only corrects the RSSM's latent state when the ball is observed. Building on RLPD, we introduce RSSM-RLPD and apply it to the task of a NAO robot learning to dribble a soccer ball into a goal. We show that the robot learns to score from multiple starting positions after 30 minutes of real-world interaction, reaching 100% success against 0% for a feed-forward policy under the same training budget. We complement our physical robot experiments with a simulation study and show that RSSM-RLPD learns policies that achieve a higher success rate than behavior cloning, a feed-forward, and GRU baseline. Taken together, our results show that the addition of a learned state representation to existing on-robot learning methods enables RL from real-robot data even with a high degree of partial observability." }, "2609.27269.pdf": { "id": "2609.27269v1", "title": "Banana Kick: Response-Informed Skill Evolution for Humanoid Soccer", "year": "2026", "authors": [ "Hao E. Zhang", "Ruize Geng", "Raihan Haque", "Khalil Zbiss", "Guanyang Luo", "Hui-ping Wang", "H. Eric Tseng", "Ding Zhao" ], "categories": [ "cs.RO" ], "pdf_url": "https://arxiv.org/pdf/2609.27269v1", "abstract": "Humanoid kicking requires coordinated whole-body motion and precise contact, while a banana kick demands contact mechanics that generate ball spin and aerodynamic curvature. Motion imitation provides a reliable ordinary-kick prior, but reinforcement learning may improve shot speed and placement accuracy without changing the underlying kicking technique. Adapting this prior to a qualitatively different contact-rich skill can fail even when the reward is dense and optimization remains stable. The failure occurs when the task objective is locally flat over the current policy's responses. We term this condition first-order learning starvation. To address it, we propose response-informed skill evolution (RISE), a closed-loop objective-continuation method for policy adaptation. RISE ranks bounded objective changes using response sensitivity estimated from cached rollouts and accepts updates only when they produce verified response progress while preserving kicking reliability. Our analysis shows that rescaling a saturated spin reward cannot recover first-order sensitivity at zero spin, whereas adapting coupled contact responses can provide a learnable path to spin generation. We integrate RISE into a humanoid kicking pipeline under calibrated contact and Magnus-force aerodynamics, and sim-to-real transfer. Experiments show that RISE evolves the ordinary kick into a high-spin curved kick with 11.55 rad/s mean ball spin, improves the mean evaluation score by 19.8% over a learning-progress curriculum, and raises joint target attainment from 15.2% to 50.9%. Ablations and response diagnostics support the mechanism, while 30 motion-capture-recorded physical trials demonstrate consistent hardware transfer of the learned curved kick. Project website: https://haozhang-thu.github.io/bananakick/" }, "2609.21100.pdf": { "id": "2609.21100v1", "title": "Dynamics-Induced Commitment in Learning-Based Robotic Penalty Kicks", "year": "2026", "authors": [ "Ruize Geng", "Hao E. Zhang", "Yisen Li", "Yikai Wang", "H. Eric Tseng", "Ding Zhao" ], "categories": [ "cs.RO" ], "pdf_url": "https://arxiv.org/pdf/2609.21100v1", "abstract": "Learning in robotic games is constrained not only by strategic information but also by what the body can still execute. We study this coupling in a hierarchical humanoid-quadruped penalty system in which game-level self-play policies command fixed soccer whole-body controllers (S-WBCs). The humanoid shooting skill is initialized from self-collected motion-capture data, whereas the quadruped saving skill is learned by reinforcement learning. We introduce dynamics-induced commitment mapping (DIC-Map), a body-grounded analysis that estimates continuation capability, identifies the first persistent loss of a terminal alternative, and tests whether the remaining interaction admits a reduced zero-sum game. For symmetric terminal alternatives, the reduced game yields a closed-form bound on optimal strategy concentration determined by the responder's value of deferring. We further show that, when the responder acts through an estimator, equal response values eliminate the direct terminal-allocation gradient and leave an estimator-mediated first-order learning channel. Experiments locate commitment about 0.29 s before contact, and changing only ball speed shifts deferral coverage. Across four responder policies, replacing the estimator raises save rate from 0.240 to 0.472, whereas a comparable gain in read accuracy obtained by waiting raises it only to 0.246. Posterior analysis is used for the equilibrium comparison because the available coverage terms are observational proxies. Project website: https://chris-ruizegeng.github.io/penaltykick/" }, "2609.05756.pdf": { "id": "2609.05756v1", "title": "Concord: A Video Relational Algebra for Cross-Modal Query Optimization", "year": "2026", "authors": [ "Sultan Muratbek", "Charisse Ivana Yeung", "Chanwut Kittivorawong", "Alvin Cheung" ], "categories": [ "cs.DB", "cs.AI" ], "pdf_url": "https://arxiv.org/pdf/2609.05756v1", "abstract": "Semantic video queries let users embed natural language prompts and use multimodal large language models (MLLMs) to interpret the video. Such queries are increasingly popular for querying video data. However, their expressiveness comes at a steep cost: an MLLM may process hours of media to return only seconds of relevant output, making naive execution slow, expensive, and inaccurate. We propose Concord, a system for expressing and optimizing semantic video queries. We makes three contributions. First, we introduce Video Relational Algebra (VRA), a nested algebra over videos, transcripts, frames, and object tracks that captures common semantic video operations. Second, we derive a set of approximate optimizations that rewrite VRA queries to reduce MLLM usage while improving result quality. For narrated video, Concord either processes transcripts instead of video or uses them to identify video clips for MLLM processing. For cross-camera queries without narration, detection and tracking replace a whole-video MLLM join with a track-level relational join. Third, we evaluate Concord on real-world videos. Across 4.59 hours of soccer broadcasts and 3.92 hours of lectures, transcript-to-video queries send only 5.32% and 2.47% of source-video duration to the MLLM and reduce MLLM cost by up to 87%. In two five-second highway clips with 18 manually adjudicated cross-camera vehicles, a Detect-Track-Join query improves F1 from .364 to .813 while making no MLLM calls. See our project at https://concord-db.github.io/." }, "2607.14809.pdf": { "id": "2607.14809v1", "title": "An LLM-Based Automatic Sportscast Solution for Robot Soccer Matches", "year": "2026", "authors": [ "Francesco Petri", "Michele Brienza", "Daniele Nardi", "Domenico Daniele Bloisi", "Aldo Gangemi", "Vincenzo Suriani" ], "categories": [ "cs.CV" ], "pdf_url": "https://arxiv.org/pdf/2607.14809v1", "abstract": "RoboCup has always been a scenario to develop systems that solve real-world problems. Driven by the main goal of playing against the 2050 FIFA World Cup champions, the RoboCup Soccer leagues need to constantly measure how the research community is progressing. Computing visual statistics from match videos is a crucial way to track this evolution. To address this challenge, this paper introduces a fully autonomous, real-time sports commentator for RoboCup matches. By bridging the gap between raw kinematic tracking and natural language generation, our neuro-symbolic architecture extracts precise statistics from video streams and turns them into fluent, hallucination-free narration. The proposed system is capable of generating statistics and commentary both during live match streaming and in post-game analysis, easily adapting to the new dynamism of the league where different humanoid robots of different sizes share the field. Supplemental materials are available at https://lab-rococo-sapienza.github.io/MARIO/" }, "2607.12702.pdf": { "id": "2607.12702v1", "title": "Vision-Based Dribbling for Humanoid Soccer via Privileged Representation Learning", "year": "2026", "authors": [ "Flavio Maiorana", "Valerio Spagnoli", "Eugenio Bugli", "Flavio Volpi", "Daniele Affinita", "Vincenzo Suriani", "Daniele Nardi", "Luca Iocchi" ], "categories": [ "cs.RO" ], "pdf_url": "https://arxiv.org/pdf/2607.12702v1", "abstract": "Recent advances in humanoid robotics have highlighted the importance of deployable loco-manipulation skills. Dribbling a soccer ball while evading active opponents requires simultaneous balance, precise ball control, and awareness of a dynamic adversary under onboard sensing and real-time constraints. Existing approaches typically separate perception and motion, which can be effective in controlled settings but may fail under occlusions, fast ball movements, and complex opponent interactions, since perception is not directly optimized for control. We propose an integrated approach in which a temporal depth encoder is embedded into a reinforcement learning policy through a task-specific projection layer. We apply this framework to a simulated Booster T1 humanoid robot and show that it is possible to learn vision-based, opponent-aware dribbling directly from depth observations, without explicit state estimation or privileged scene information. The learned policy achieves 100% success in nominal target-driven dribbling and 96% success with a single static obstacle, while reaching 46% success against an actively moving ball-attacker opponent. These results demonstrate that the proposed framework supports robust vision-based dribbling in nominal and moderately dynamic settings, and provides a strong foundation for handling more challenging moving-adversary scenarios." }, "2606.17388.pdf": { "id": "2606.17388v1", "title": "Agent Utilities over Generalized Voronoi Regions and their Gradients", "year": "2026", "authors": [ "Andre N. Costa", "Petter \u00d6gren", "Carlos H. C. Ribeiro" ], "categories": [ "cs.RO", "cs.CG", "eess.SY" ], "pdf_url": "https://arxiv.org/pdf/2606.17388v1", "abstract": "In this paper, we generalize the concept of Voronoi regions, define agent utility as the integral of a utility density over the corresponding Voronoi region, derive gradients of the utility, and illustrate the approach in a two-team example from soccer. The generalization of Voronoi regions is in the form of so-called Cost-Induced Voronoi (CIV) regions, where the agent state space may differ from the space being partitioned. One example of such regions is when the cost is given by the optimal solution of an LQR control problem. Then the agent states include position as well as velocity, while the partitioned space only includes positions. The agent utility is defined by integrating some utility density over the CIV region of the agent. This utility density might be the probability density of some beneficial event, such as receiving a pass in soccer. The utility is then the overall probability of receiving a pass and the gradient represents a way to improve that probability. We show how this utility gradient can be computed using the Reynolds Transport Theorem from fluid mechanics, and that this approach achieves similar accuracy while reducing computation time by about an order of magnitude compared to a baseline finite-difference approximation." }, "2606.11092.pdf": { "id": "2606.11092v3", "title": "RoboNaldo: Accurate, Stable and Powerful Humanoid Soccer Shooting via Motion-Guided Curriculum Reinforcement Learning", "year": "2026", "authors": [ "Yichao Zhong", "Yidan Lu", "Yuhang Lu", "Tianyang Tang", "Haoguang Mai", "Yixuan Pan", "Tianyu Li", "Li Chen", "Jingbo Wang", "Zhongyu Li", "Peng Lu", "Hongyang Li" ], "categories": [ "cs.RO", "cs.AI" ], "pdf_url": "https://arxiv.org/pdf/2606.11092v3", "abstract": "Elite humanoid soccer shooting requires whole-body stability, high-impulse whole-body interactions, and accuracy to targets. Motion tracking-driven reinforcement learning (RL) provides stability in whole-body movement coordination, but a fixed reference makes it hard to adapt to varied ball positions and strike timings; in contrast, task reward-driven RL struggles to explore and discover valid kicks from scratch. We therefore introduce RoboNaldo, a three-stage motion-guided curriculum RL framework for high-impulse humanoid interaction. A single human-kick reference is used as a scaffold and progressively shifts optimization towards shooting performance. The curriculum first learns a stable whole-body kicking prior, then adapts the kick to free-kick settings where the ball is stationary at random positions, and finally extends it to moving-ball shooting through a locomotion-command and kick-trigger interface. A high-level heuristic planner controls this interface during training, while alternative high-level controllers can drive the same low-level policy at inference. In simulation, RoboNaldo demonstrates free-kick shot error 48.6% lower and shoot velocity 2.96x than prior work baselines. In real world on a Unitree G1 with onboard perception, RoboNaldo attains 0.73 m and 0.86 m average target shooting error from 3 m away in free-kick and moving-ball cases, accordingly. And the post-contact ball velocity reaches 13.10 m/s, which is 59-71% of reported professional open-play shot speed. Project page: https://opendrivelab.com/RoboNaldo." }, "2604.14534.pdf": { "id": "2604.14534v1", "title": "An unsupervised decision-support framework for multivariate biomarker analysis in athlete monitoring", "year": "2026", "authors": [ "Fernando Barcelos Rosito", "Sebasti\u00e3o De Jesus Menezes", "Simone Ferreira Sturza", "Adriana Seixas", "Muriel Figueredo Franco" ], "categories": [ "cs.LG", "stat.AP" ], "pdf_url": "https://arxiv.org/pdf/2604.14534v1", "abstract": "Purpose. Athlete monitoring is constrained by small cohorts, heterogeneous biomarker scales, limited feasibility of repeated sampling, and the lack of reliable injury ground truth. These limitations reduce the interpretability and utility of traditional univariate and binary risk models. This study addresses these challenges by proposing an unsupervised multivariate framework to identify latent physiological states in athletes using real data. Methods. We propose a modular computational framework that operates in the joint biomarker space, integrating preprocessing, clinical safety screening, unsupervised clustering, and centroid-based physiological interpretation. Profiles are learned exclusively from amateur soccer players during a competitive microcycle. Synthetic data augmentation evaluates robustness and scalability. Ward hierarchical clustering supports monitoring and etiological differentiation, while Gaussian Mixture Models (GMM) enable structural stability analysis in high-dimensional settings. Results. The framework identifies coherent profiles that distinguish mechanical damage from metabolic stress while preserving homeostatic states. Synthetic data augmentation demonstrates feasibility and detection of latent silent risk phenotypes typically missed by univariate monitoring. Structural analyses indicate robustness under augmentation and higher-dimensional settings. Conclusion. The framework enables interpretable identification of latent physiological states from multivariate biomarker data without injury labels. By distinguishing mechanisms and revealing silent risk patterns not captured by conventional monitoring, it provides actionable insights for individualized athlete monitoring and decision making." }, "2512.22596.pdf": { "id": "2512.22596v3", "title": "Evaluating Soccer Player Movements Using the Attacker-Defender Model", "year": "2025", "authors": [ "Takuma Narizuka", "Issei Yamazaki" ], "categories": [ "physics.soc-ph" ], "pdf_url": "https://arxiv.org/pdf/2512.22596v3", "abstract": "The present study investigates the attacker-defender (AD) model proposed by Brink et al. (2023), a motion model that describes the interactions between a ball carrier (attacker) and the nearest defender during ball possession. The model is based on the equations of motion for both players, incorporating resistance, goal-oriented force, and opponent-oriented force. It generates trajectories based on physically interpretable parameters. Although the AD model reproduces real dribbling trajectories well, previous studies have explored only a limited range of parameter values and relied on relatively small datasets. This study aims to (1) enhance parameter optimization by solving the AD model for one player with the opponent's actual trajectory fixed, (2) validate the model's applicability to a large dataset from 306 J1 League matches, and (3) demonstrate distinct playing styles of attackers and defenders based on the full range of optimized parameters." }, "2512.12437.pdf": { "id": "2512.12437v1", "title": "Sim2Real Reinforcement Learning for Soccer skills", "year": "2025", "authors": [ "Jonathan Spraggett" ], "categories": [ "cs.RO", "cs.LG" ], "pdf_url": "https://arxiv.org/pdf/2512.12437v1", "abstract": "This thesis work presents a more efficient and effective approach to training control-related tasks for humanoid robots using Reinforcement Learning (RL). The traditional RL methods are limited in adapting to real-world environments, complexity, and natural motions, but the proposed approach overcomes these limitations by using curriculum training and Adversarial Motion Priors (AMP) technique. The results show that the developed RL policies for kicking, walking, and jumping are more dynamic, and adaptive, and outperformed previous methods. However, the transfer of the learned policy from simulation to the real world was unsuccessful, highlighting the limitations of current RL methods in fully adapting to real-world scenarios." }, "2511.17064.pdf": { "id": "2511.17064v1", "title": "Single-Dataset Meta-Analysis For Many-Analysts And Multiverse Studies", "year": "2025", "authors": [ "Franti\u0161ek Barto\u0161", "Suzanne Hoogeveen", "Alexandra Sarafoglou", "Samuel Pawel" ], "categories": [ "stat.ME" ], "pdf_url": "https://arxiv.org/pdf/2511.17064v1", "abstract": "Empirical claims often rely on one population, design, and analysis. Many-analysts, multiverse, and robustness studies expose how results can vary across plausible analytic choices. Synthesizing these results, however, is nontrivial as all results are computed from the same dataset. We introduce single-dataset meta-analysis, a weighted-likelihood approach that incorporates the information in the dataset at most once. It prevents overconfident inferences that would arise if a standard meta-analysis was applied to the data. Single-dataset meta-analysis yields meta-analytic point and interval estimates of the average effect across analytic approaches and of between-analyst heterogeneity, and can be supplied by classical and Bayesian hypothesis tests. Both the common-effect and random-effects versions of the model can be estimated by standard meta-analytic software with small input adjustments. We demonstrate the method via application to the many-analysts study on racial bias in soccer, the many-analysts study of marital status and cardiovascular disease, and the multiverse study on technology use and well-being. The results show how single-dataset meta-analysis complements the qualitative evaluation of many-analysts and multiverse studies." }, "2511.09455.pdf": { "id": "2511.09455v1", "title": "Hand Held Multi-Object Tracking Dataset in American Football", "year": "2025", "authors": [ "Rintaro Otsubo", "Kanta Sawafuji", "Hideo Saito" ], "categories": [ "cs.CV" ], "pdf_url": "https://arxiv.org/pdf/2511.09455v1", "abstract": "Multi-Object Tracking (MOT) plays a critical role in analyzing player behavior from videos, enabling performance evaluation. Current MOT methods are often evaluated using publicly available datasets. However, most of these focus on everyday scenarios such as pedestrian tracking or are tailored to specific sports, including soccer and basketball. Despite the inherent challenges of tracking players in American football, such as frequent occlusion and physical contact, no standardized dataset has been publicly available, making fair comparisons between methods difficult. To address this gap, we constructed the first dedicated detection and tracking dataset for the American football players and conducted a comparative evaluation of various detection and tracking methods. Our results demonstrate that accurate detection and tracking can be achieved even in crowded scenarios. Fine-tuning detection models improved performance over pre-trained models. Furthermore, when these fine-tuned detectors and re-identification models were integrated into tracking systems, we observed notable improvements in tracking accuracy compared to existing approaches. This work thus enables robust detection and tracking of American football players in challenging, high-density scenarios previously underserved by conventional methods." }, "2509.13605.pdf": { "id": "2509.13605v1", "title": "A Generalization of CLAP from 3D Localization to Image Processing, A Connection With RANSAC & Hough Transforms", "year": "2025", "authors": [ "Ruochen Hou", "Gabriel I. Fernandez", "Alex Xu", "Dennis W. Hong" ], "categories": [ "cs.CV", "cs.RO" ], "pdf_url": "https://arxiv.org/pdf/2509.13605v1", "abstract": "In previous work, we introduced a 2D localization algorithm called CLAP, Clustering to Localize Across $n$ Possibilities, which was used during our championship win in RoboCup 2024, an international autonomous humanoid soccer competition. CLAP is particularly recognized for its robustness against outliers, where clustering is employed to suppress noise and mitigate against erroneous feature matches. This clustering-based strategy provides an alternative to traditional outlier rejection schemes such as RANSAC, in which candidates are validated by reprojection error across all data points. In this paper, CLAP is extended to a more general framework beyond 2D localization, specifically to 3D localization and image stitching. We also show how CLAP, RANSAC, and Hough transforms are related. The generalization of CLAP is widely applicable to many different fields and can be a useful tool to deal with noise and uncertainty." }, "2509.08495.pdf": { "id": "2509.08495v1", "title": "CLAP: Clustering to Localize Across n Possibilities, A Simple, Robust Geometric Approach in the Presence of Symmetries", "year": "2025", "authors": [ "Gabriel I. Fernandez", "Ruochen Hou", "Alex Xu", "Colin Togashi", "Dennis W. Hong" ], "categories": [ "cs.RO" ], "pdf_url": "https://arxiv.org/pdf/2509.08495v1", "abstract": "In this paper, we present our localization method called CLAP, Clustering to Localize Across $n$ Possibilities, which helped us win the RoboCup 2024 adult-sized autonomous humanoid soccer competition. Competition rules limited our sensor suite to stereo vision and an inertial sensor, similar to humans. In addition, our robot had to deal with varying lighting conditions, dynamic feature occlusions, noise from high-impact stepping, and mistaken features from bystanders and neighboring fields. Therefore, we needed an accurate, and most importantly robust localization algorithm that would be the foundation for our path-planning and game-strategy algorithms. CLAP achieves these requirements by clustering estimated states of our robot from pairs of field features to localize its global position and orientation. Correct state estimates naturally cluster together, while incorrect estimates spread apart, making CLAP resilient to noise and incorrect inputs. CLAP is paired with a particle filter and an extended Kalman filter to improve consistency and smoothness. Tests of CLAP with other landmark-based localization methods showed similar accuracy. However, tests with increased false positive feature detection showed that CLAP outperformed other methods in terms of robustness with very little divergence and velocity jumps. Our localization performed well in competition, allowing our robot to shoot faraway goals and narrowly defend our goal." }, "2509.01439.pdf": { "id": "2509.01439v1", "title": "SoccerHigh: A Benchmark Dataset for Automatic Soccer Video Summarization", "year": "2025", "authors": [ "Artur D\u00edaz-Juan", "Coloma Ballester", "Gloria Haro" ], "categories": [ "cs.CV", "cs.AI", "cs.MM" ], "pdf_url": "https://arxiv.org/pdf/2509.01439v1", "abstract": "Video summarization aims to extract key shots from longer videos to produce concise and informative summaries. One of its most common applications is in sports, where highlight reels capture the most important moments of a game, along with notable reactions and specific contextual events. Automatic summary generation can support video editors in the sports media industry by reducing the time and effort required to identify key segments. However, the lack of publicly available datasets poses a challenge in developing robust models for sports highlight generation. In this paper, we address this gap by introducing a curated dataset for soccer video summarization, designed to serve as a benchmark for the task. The dataset includes shot boundaries for 237 matches from the Spanish, French, and Italian leagues, using broadcast footage sourced from the SoccerNet dataset. Alongside the dataset, we propose a baseline model specifically designed for this task, which achieves an F1 score of 0.3956 in the test set. Furthermore, we propose a new metric constrained by the length of each target summary, enabling a more objective evaluation of the generated content. The dataset and code are available at https://ipcv.github.io/SoccerHigh/." }, "2508.11569.pdf": { "id": "2508.11569v1", "title": "TrajSV: A Trajectory-based Model for Sports Video Representations and Applications", "year": "2025", "authors": [ "Zheng Wang", "Shihao Xu", "Wei Shi" ], "categories": [ "cs.CV", "cs.IR" ], "pdf_url": "https://arxiv.org/pdf/2508.11569v1", "abstract": "Sports analytics has received significant attention from both academia and industry in recent years. Despite the growing interest and efforts in this field, several issues remain unresolved, including (1) data unavailability, (2) lack of an effective trajectory-based framework, and (3) requirement for sufficient supervision labels. In this paper, we present TrajSV, a trajectory-based framework that addresses various issues in existing studies. TrajSV comprises three components: data preprocessing, Clip Representation Network (CRNet), and Video Representation Network (VRNet). The data preprocessing module extracts player and ball trajectories from sports broadcast videos. CRNet utilizes a trajectory-enhanced Transformer module to learn clip representations based on these trajectories. Additionally, VRNet learns video representations by aggregating clip representations and visual features with an encoder-decoder architecture. Finally, a triple contrastive loss is introduced to optimize both video and clip representations in an unsupervised manner. The experiments are conducted on three broadcast video datasets to verify the effectiveness of TrajSV for three types of sports (i.e., soccer, basketball, and volleyball) with three downstream applications (i.e., sports video retrieval, action spotting, and video captioning). The results demonstrate that TrajSV achieves state-of-the-art performance in sports video retrieval, showcasing a nearly 70% improvement. It outperforms baselines in action spotting, achieving state-of-the-art results in 9 out of 17 action categories, and demonstrates a nearly 20% improvement in video captioning. Additionally, we introduce a deployed system along with the three applications based on TrajSV." }, "2508.08290.pdf": { "id": "2508.08290v2", "title": "Evaluating Competitiveness in UEFA's New Champions League Format", "year": "2025", "authors": [ "Karel Devriesere", "Dries Goossens", "Frits Spieksma" ], "categories": [ "physics.soc-ph" ], "pdf_url": "https://arxiv.org/pdf/2508.08290v2", "abstract": "Recently, UEFA changed the group stage of its international soccer competitions to an incomplete round robin tournament. Previously, teams were divided into groups, each playing a double round robin tournament with a resulting ranking table. In contrast, the new format has all teams competing in one league, producing a single ranking. We investigate the effect of the new format on the number of competitive matches in the UEFA Champions League. A match is non-competitive if the prize for at least one opponent does not depend on the match outcome, or if there exists an opportunity for both opponents to collude; otherwise, we call a match competitive. Using Monte Carlo simulations, we show that the new format results in more competitive matches than the old format." }, "2506.16821.pdf": { "id": "2506.16821v1", "title": "Self-supervised Feature Extraction for Enhanced Ball Detection on Soccer Robots", "year": "2025", "authors": [ "Can Lin", "Daniele Affinita", "Marco E. P. Zimmatore", "Daniele Nardi", "Domenico D. Bloisi", "Vincenzo Suriani" ], "categories": [ "cs.CV" ], "pdf_url": "https://arxiv.org/pdf/2506.16821v1", "abstract": "Robust and accurate ball detection is a critical component for autonomous humanoid soccer robots, particularly in dynamic and challenging environments such as RoboCup outdoor fields. However, traditional supervised approaches require extensive manual annotation, which is costly and time-intensive. To overcome this problem, we present a self-supervised learning framework for domain-adaptive feature extraction to enhance ball detection performance. The proposed approach leverages a general-purpose pretrained model to generate pseudo-labels, which are then used in a suite of self-supervised pretext tasks -- including colorization, edge detection, and triplet loss -- to learn robust visual features without relying on manual annotations. Additionally, a model-agnostic meta-learning (MAML) strategy is incorporated to ensure rapid adaptation to new deployment scenarios with minimal supervision. A new dataset comprising 10,000 labeled images from outdoor RoboCup SPL matches is introduced, used to validate the method, and made available to the community. Experimental results demonstrate that the proposed pipeline outperforms baseline models in terms of accuracy, F1 score, and IoU, while also exhibiting faster convergence." }, "2506.11934.pdf": { "id": "2506.11934v1", "title": "Temporal Dynamics of Emotions in Italian Online Soccer Fandoms", "year": "2025", "authors": [ "Salvatore Citraro", "Giovanni Mauro", "Emanuele Ferragina" ], "categories": [ "cs.SI", "cs.MM" ], "pdf_url": "https://arxiv.org/pdf/2506.11934v1", "abstract": "This study investigates the emotional dynamics of Italian soccer fandoms through computational analysis of user-generated content from official Instagram accounts of 83 teams across Serie A, Serie B, and Lega Pro during the 2023-24 season. By applying sentiment analysis to fan comments, we extract temporal emotional patterns and identify distinct clusters of fan bases with similar preseason expectations. Drawing from complex systems theory, we characterize joy as displaying anti-bursty temporal distributions, while anger is marked by pronounced bursty patterns. Our analysis reveals significant correlations between these emotional signals, preseason expectations, socioeconomic factors, and final league rankings. In particular, the burstiness metric emerges as a meaningful correlate of team performance; statistical models excluding this parameter show a decrease in the coefficient of determination of 32%. These findings offer novel insights into the relationship between fan emotional expression and team outcomes, suggesting potential avenues for research in sports analytics, social media dynamics, and fan engagement studies." }, "2505.24629.pdf": { "id": "2505.24629v1", "title": "Stop Guessing: Optimizing Goalkeeper Policies for Soccer Penalty Kicks", "year": "2025", "authors": [ "Lotte Bransen", "Tim Janssen", "Jesse Davis" ], "categories": [ "cs.LG", "cs.GT" ], "pdf_url": "https://arxiv.org/pdf/2505.24629v1", "abstract": "Penalties are fraught and game-changing moments in soccer games that teams explicitly prepare for. Consequently, there has been substantial interest in analyzing them in order to provide advice to practitioners. From a data science perspective, such analyses suffer from a significant limitation: they make the unrealistic simplifying assumption that goalkeepers and takers select their action -- where to dive and where to the place the kick -- independently of each other. In reality, the choices that some goalkeepers make depend on the taker's movements and vice-versa. This adds substantial complexity to the problem because not all players have the same action capacities, that is, only some players are capable of basing their decisions on their opponent's movements. However, the small sample sizes on the player level mean that one may have limited insights into a specific opponent's capacities. We address these challenges by developing a player-agnostic simulation framework that can evaluate the efficacy of different goalkeeper strategies. It considers a rich set of choices and incorporates information about a goalkeeper's skills. Our work is grounded in a large dataset of penalties that were annotated by penalty experts and include aspects of both kicker and goalkeeper strategies. We show how our framework can be used to optimize goalkeeper policies in real-world situations." }, "2503.08945.pdf": { "id": "2503.08945v1", "title": "PassAI: explainable artificial intelligence algorithm for soccer pass analysis using multimodal information resources", "year": "2025", "authors": [ "Ryota Takamido", "Jun Ota", "Hiroki Nakamoto" ], "categories": [ "cs.HC" ], "pdf_url": "https://arxiv.org/pdf/2503.08945v1", "abstract": "This study developed a new explainable artificial intelligence algorithm called PassAI, which classifies successful or failed passes in a soccer game and explains its rationale using both tracking and passer's seasonal stats information. This study aimed to address two primary challenges faced by artificial intelligence and machine learning algorithms in the sports domain: how to use different modality data for the analysis and how to explain the rationale of the outcome from multimodal perspectives. To address these challenges, PassAI has two processing streams for multimodal information: tracking image data and passer's stats and classifying pass success and failure. After completing the classification, it provides a rationale by either calculating the relative contribution between the different modality data or providing more detailed contribution factors within the modality. The results of the experiment with 6,349 passes of data obtained from professional soccer games revealed that PassAI showed higher classification performance than state-of-the-art algorithms by >5% and could visualize the rationale of the pass success/failure for both tracking and stats data. These results highlight the importance of using multimodality data in the sports domain to increase the performance of the artificial intelligence algorithm and explainability of the outcomes." }, "2505.15820.pdf": { "id": "2505.15820v4", "title": "Common Data Format (CDF): A Standardized Format for Match-Data in Football (Soccer)", "year": "2025", "authors": [ "Gabriel Anzer", "Kilian Arnsmeyer", "Pascal Bauer", "Joris Bekkers", "Ulf Brefeld", "Jesse Davis", "Nicolas Evans", "Matthias Kempe", "Samuel J Robertson", "Joshua Wyatt Smith", "Jan Van Haaren" ], "categories": [ "cs.DB", "cs.AI" ], "pdf_url": "https://arxiv.org/pdf/2505.15820v4", "abstract": "During football matches, a variety of different parties (e.g., companies) each collect (possibly overlapping) data about the match ranging from basic information (e.g., starting players) to detailed positional data. This data is provided to clubs, federations, and other organizations who are increasingly interested in leveraging this data to inform their decision making. Unfortunately, analyzing such data pose significant barriers because each provider may (1) collect different data, (2) use different specifications even within the same category of data, (3) represent the data differently, and (4) delivers the data in a different manner (e.g., file format, protocol). Consequently, working with these data requires a significant investment of time and money. The goal of this work is to propose a uniform and standardized format for football data called the Common Data Format (CDF). The CDF specifies a minimal schema for five types of match data: match sheet data, video footage, event data, tracking data, and match meta data. It aims to ensure that the provided data is clear, sufficiently contextualized (e.g., its provenance is clear), and complete such that it enables common downstream analysis tasks. Concretely, this paper will detail the technical specifications of the CDF, the representational choices that were made to help ensure the clarity of the provided data, and a concrete approach for delivering data in the CDF. This represents Version 1.0.0 of the CDF." }, "2405.18565.pdf": { "id": "2405.18565v1", "title": "Video2MR: Automatically Generating Mixed Reality 3D Instructions by Augmenting Extracted Motion from 2D Videos", "year": "2024", "authors": [ "Keiichi Ihara", "Kyzyl Monteiro", "Mehrad Faridan", "Rubaiat Habib Kazi", "Ryo Suzuki" ], "categories": [ "cs.HC" ], "pdf_url": "https://arxiv.org/pdf/2405.18565v1", "abstract": "This paper introduces Video2MR, a mixed reality system that automatically generates 3D sports and exercise instructions from 2D videos. Mixed reality instructions have great potential for physical training, but existing works require substantial time and cost to create these 3D experiences. Video2MR overcomes this limitation by transforming arbitrary instructional videos available online into MR 3D avatars with AI-enabled motion capture (DeepMotion). Then, it automatically enhances the avatar motion through the following augmentation techniques: 1) contrasting and highlighting differences between the user and avatar postures, 2) visualizing key trajectories and movements of specific body parts, 3) manipulation of time and speed using body motion, and 4) spatially repositioning avatars for different perspectives. Developed on Hololens 2 and Azure Kinect, we showcase various use cases, including yoga, dancing, soccer, tennis, and other physical exercises. The study results confirm that Video2MR provides more engaging and playful learning experiences, compared to existing 2D video instructions." }, "2401.15026.pdf": { "id": "2401.15026v1", "title": "Multi-Agent Coordination for a Partially Observable and Dynamic Robot Soccer Environment with Limited Communication", "year": "2024", "authors": [ "Daniele Affinita", "Flavio Volpi", "Valerio Spagnoli", "Vincenzo Suriani", "Daniele Nardi", "Domenico D. Bloisi" ], "categories": [ "cs.RO", "cs.MA" ], "pdf_url": "https://arxiv.org/pdf/2401.15026v1", "abstract": "RoboCup represents an International testbed for advancing research in AI and robotics, focusing on a definite goal: developing a robot team that can win against the human world soccer champion team by the year 2050. To achieve this goal, autonomous humanoid robots' coordination is crucial. This paper explores novel solutions within the RoboCup Standard Platform League (SPL), where a reduction in WiFi communication is imperative, leading to the development of new coordination paradigms. The SPL has experienced a substantial decrease in network packet rate, compelling the need for advanced coordination architectures to maintain optimal team functionality in dynamic environments. Inspired by market-based task assignment, we introduce a novel distributed coordination system to orchestrate autonomous robots' actions efficiently in low communication scenarios. This approach has been tested with NAO robots during official RoboCup competitions and in the SimRobot simulator, demonstrating a notable reduction in task overlaps in limited communication settings." }, "2310.17058.pdf": { "id": "2310.17058v1", "title": "Diseno y Desarrollo de Prototipos Roboticos para Competencias de Futbol utilizando Motores Dynamixel", "year": "2023", "authors": [ "Pablo Moraes", "Hiago Sodre", "Monica Rodriguez", "Andre Kelbouscas", "Jean Schuster", "Cristiano Schuster", "Ricardo Grando" ], "categories": [ "cs.RO" ], "pdf_url": "https://arxiv.org/pdf/2310.17058v1", "abstract": "This article describes the design and development of robotic prototypes for robotic soccer competitions using Dynamixel motors. Although the prototypes are not aimed at world-class competitions, they represent a significant step in the development of sports robots. Model XL430-W250 Dynamixel motors were chosen and electronic circuits were implemented using control boards such as OpenCR and Raspberry Pi 3. A crucial component was introduced: a step-up board that charges a capacitor to create a powerful kick to the ball via anelectromagnet controlled by Arduino Nano. The programming and coordination of the prototypes was carried out using the ROS environment (Robot Operating System), which allows effective integration of movements and communication. Although the prototypes were not optimized for global competition, they underwent extensive testing, evaluating their speed and maneuverability, as well as soccer tactics in the GRSim simulator. These prototypes contribute to the further development of sports robotics and illustrate the research potential in this exciting area." }, "2304.03754.pdf": { "id": "2304.03754v1", "title": "Language Models are Causal Knowledge Extractors for Zero-shot Video Question Answering", "year": "2023", "authors": [ "Hung-Ting Su", "Yulei Niu", "Xudong Lin", "Winston H. Hsu", "Shih-Fu Chang" ], "categories": [ "cs.CL", "cs.CV" ], "pdf_url": "https://arxiv.org/pdf/2304.03754v1", "abstract": "Causal Video Question Answering (CVidQA) queries not only association or temporal relations but also causal relations in a video. Existing question synthesis methods pre-trained question generation (QG) systems on reading comprehension datasets with text descriptions as inputs. However, QG models only learn to ask association questions (e.g., ``what is someone doing...'') and result in inferior performance due to the poor transfer of association knowledge to CVidQA, which focuses on causal questions like ``why is someone doing ...''. Observing this, we proposed to exploit causal knowledge to generate question-answer pairs, and proposed a novel framework, Causal Knowledge Extraction from Language Models (CaKE-LM), leveraging causal commonsense knowledge from language models to tackle CVidQA. To extract knowledge from LMs, CaKE-LM generates causal questions containing two events with one triggering another (e.g., ``score a goal'' triggers ``soccer player kicking ball'') by prompting LM with the action (soccer player kicking ball) to retrieve the intention (to score a goal). CaKE-LM significantly outperforms conventional methods by 4% to 6% of zero-shot CVidQA accuracy on NExT-QA and Causal-VidQA datasets. We also conduct comprehensive analyses and provide key findings for future research." }, "2302.10817.pdf": { "id": "2302.10817v1", "title": "Inferring Implicit Trait Preferences for Task Allocation in Heterogeneous Teams", "year": "2023", "authors": [ "Vivek Mallampati", "Harish Ravichandar" ], "categories": [ "cs.RO", "cs.MA" ], "pdf_url": "https://arxiv.org/pdf/2302.10817v1", "abstract": "Task allocation in heterogeneous multi-agent teams often requires reasoning about multi-dimensional agent traits (i.e., capabilities) and the demands placed on them by tasks. However, existing methods tend to ignore the fact that not all traits equally contribute to a given task. Ignoring such inherent preferences or relative importance can lead to unintended sub-optimal allocations of limited agent resources that do not necessarily contribute to task success. Further, reasoning over a large number of traits can incur a hefty computational burden. To alleviate these concerns, we propose an algorithm to infer task-specific trait preferences implicit in expert demonstrations. We leverage the insight that the consistency with which an expert allocates a trait to a task across demonstrations reflects the trait's importance to that task. Inspired by findings in psychology, we account for the fact that the inherent diversity of a trait in the dataset influences the dataset's informativeness and, thereby, the extent of the inferred preference or the lack thereof. Through detailed numerical simulations and evaluations of a publicly-available soccer dataset (FIFA 20), we demonstrate that we can successfully infer implicit trait preferences and that accounting for the inferred preferences leads to more computationally efficient and effective task allocation, compared to a baseline approach that treats all traits equally." }, "2302.08120.pdf": { "id": "2302.08120v1", "title": "Generalised Uncertainty Relations from Finite-Accuracy Measurements", "year": "2023", "authors": [ "Matthew J. Lake", "Marek Miller", "Ray Ganardi", "Tomasz Paterek" ], "categories": [ "gr-qc", "hep-ph", "quant-ph" ], "pdf_url": "https://arxiv.org/pdf/2302.08120v1", "abstract": "In this short note we show how the Generalised Uncertainty Principle (GUP) and the Extended Uncertainty Principle (EUP), two of the most common generalised uncertainty relations proposed in the quantum gravity literature, can be derived within the context of canonical quantum theory, without the need for modified commutation relations. A GUP-type relation naturally emerges when the standard position operator is replaced by an appropriate Positive Operator Valued Measure (POVM), representing a finite-accuracy measurement that localises the quantum wave packet to within a spatial region $\u03c3_g > 0$. This length scale is the standard deviation of the envelope function, $g$, that defines the POVM elements. Similarly, an EUP-type relation emerges when the standard momentum operator is replaced by a POVM that localises the wave packet to within a region $\\tilde\u03c3_g > 0$ in momentum space. The usual GUP and EUP are recovered by setting $\u03c3_g \\simeq \\sqrt{\\hbar G/c^3}$, the Planck length, and $\\tilde\u03c3_g \\simeq \\hbar\\sqrt{\u039b/3}$, where $\u039b$ is the cosmological constant. Crucially, the canonical Hamiltonian and commutation relations, and, hence, the canonical Schr{\\\" o}dinger and Heisenberg equations, remain unchanged. This demonstrates that GUP and EUP phenomenology can be obtained without modified commutators, which are known to lead to various pathologies, including violation of the equivalence principle, violation of Lorentz invariance in the relativistic limit, the reference frame-dependence of the `minimum' length, and the so-called soccer ball problem for multi-particle states." }, "2301.01349.pdf": { "id": "2301.01349v2", "title": "Quantitative Planning with Action Deception in Concurrent Stochastic Games", "year": "2023", "authors": [ "Chongyang Shi", "Shuo Han", "Jie Fu" ], "categories": [ "eess.SY" ], "pdf_url": "https://arxiv.org/pdf/2301.01349v2", "abstract": "We study a class of two-player competitive concurrent stochastic games on graphs with reachability objectives. Specifically, player 1 aims to reach a subset $F_1$ of game states, and player 2 aims to reach a subset $F_2$ of game states where $F_2\\cap F_1=\\emptyset$. Both players aim to satisfy their reachability objectives before their opponent does. Yet, the information players have about the game dynamics is asymmetric: P1 has a (set of) hidden actions unknown to P2 at the beginning of their interaction. In this setup, we investigate P1's strategic planning of action deception that decides when to deviate from the Nash equilibrium in P2's game model and employ a hidden action, so that P1 can maximize the value of action deception, which is the additional payoff compared to P1's payoff in the game where P2 has complete information. Anticipating that P2 may detect his misperception about the game and adapt his strategy during interaction in unpredictable ways, we construct a planning problem for P1 to augment the game model with an incomplete model about the theory of mind of the opponent P2. While planning in the augmented game, P1 can effectively influence P2's perception so as to entice P2 to take actions that benefit P1. We prove that the proposed deceptive planning algorithm maximizes a lower bound on the value of action deception and demonstrate the effectiveness of our deceptive planning algorithm using a robot motion planning problem inspired by soccer games." }, "2201.04024.pdf": { "id": "2201.04024v1", "title": "Smart Director: An Event-Driven Directing System for Live Broadcasting", "year": "2022", "authors": [ "Yingwei Pan", "Yue Chen", "Qian Bao", "Ning Zhang", "Ting Yao", "Jingen Liu", "Tao Mei" ], "categories": [ "cs.CV", "cs.MM" ], "pdf_url": "https://arxiv.org/pdf/2201.04024v1", "abstract": "Live video broadcasting normally requires a multitude of skills and expertise with domain knowledge to enable multi-camera productions. As the number of cameras keep increasing, directing a live sports broadcast has now become more complicated and challenging than ever before. The broadcast directors need to be much more concentrated, responsive, and knowledgeable, during the production. To relieve the directors from their intensive efforts, we develop an innovative automated sports broadcast directing system, called Smart Director, which aims at mimicking the typical human-in-the-loop broadcasting process to automatically create near-professional broadcasting programs in real-time by using a set of advanced multi-view video analysis algorithms. Inspired by the so-called \"three-event\" construction of sports broadcast, we build our system with an event-driven pipeline consisting of three consecutive novel components: 1) the Multi-view Event Localization to detect events by modeling multi-view correlations, 2) the Multi-view Highlight Detection to rank camera views by the visual importance for view selection, 3) the Auto-Broadcasting Scheduler to control the production of broadcasting videos. To our best knowledge, our system is the first end-to-end automated directing system for multi-camera sports broadcasting, completely driven by the semantic understanding of sports events. It is also the first system to solve the novel problem of multi-view joint event detection by cross-view relation modeling. We conduct both objective and subjective evaluations on a real-world multi-camera soccer dataset, which demonstrate the quality of our auto-generated videos is comparable to that of the human-directed. Thanks to its faster response, our system is able to capture more fast-passing and short-duration events which are usually missed by human directors." }, "2108.07057.pdf": { "id": "2108.07057v1", "title": "Data-driven Analysis of Gender Differences and Similarities in Scratch Programs", "year": "2021", "authors": [ "Isabella Gra\u00dfl", "Katharina Geldreich", "Gordon Fraser" ], "categories": [ "cs.SE" ], "pdf_url": "https://arxiv.org/pdf/2108.07057v1", "abstract": "Block-based programming environments such as Scratch are an essential entry point to computer science. In order to create an effective learning environment that has the potential to address the gender imbalance in computer science, it is essential to better understand gender-specific differences in how children use such programming environments. In this paper, we explore gender differences and similarities in Scratch programs along two dimensions: In order to understand what motivates girls and boys to use Scratch, we apply a topic analysis using unsupervised machine learning for the first time on Scratch programs, using a dataset of 317 programs created by girls and boys in the range of 8-10 years. In order to understand how they program for these topics, we apply automated program analysis on the code implemented in these projects. We find that, in-line with common stereotypes, girls prefer topics that revolve around unicorns, celebrating, dancing and music, while boys tend to prefer gloomy topics with bats and ghouls, or competitive ones such as soccer or basketball. Girls prefer animations and stories, resulting in simpler control structures, while boys create games with more loops and conditional statements, resulting in more complex programs. Considering these differences can help to improve the learning outcomes and the resulting computing-related self-concepts, which are prerequisites for developing a longer-term interest in computer science." }, "2108.06432.pdf": { "id": "2108.06432v2", "title": "Soccer line mark segmentation and classification with stochastic watershed transform", "year": "2021", "authors": [ "Daniel Berj\u00f3n", "Carlos Cuevas", "Narciso Garc\u00eda" ], "categories": [ "cs.CV" ], "pdf_url": "https://arxiv.org/pdf/2108.06432v2", "abstract": "Augmented reality applications are beginning to change the way sports are broadcast, providing richer experiences and valuable insights to fans. The first step of augmented reality systems is camera calibration, possibly based on detecting the line markings of the playing field. Most existing proposals for line detection rely on edge detection and Hough transform, but radial distortion and extraneous edges cause inaccurate or spurious detections of line markings. We propose a novel strategy to automatically and accurately segment and classify line markings. First, line points are segmented thanks to a stochastic watershed transform that is robust to radial distortions, since it makes no assumptions about line straightness, and is unaffected by the presence of players or the ball. The line points are then linked to primitive structures (straight lines and ellipses) thanks to a very efficient procedure that makes no assumptions about the number of primitives that appear in each image. The strategy has been tested on a new and public database composed by 60 annotated images from matches in five stadiums. The results obtained have proven that the proposed strategy is more robust and accurate than existing approaches, achieving successful line mark detection even in challenging conditions." }, "2108.03936.pdf": { "id": "2108.03936v1", "title": "3D Human Reconstruction in the Wild with Collaborative Aerial Cameras", "year": "2021", "authors": [ "Cherie Ho", "Andrew Jong", "Harry Freeman", "Rohan Rao", "Rogerio Bonatti", "Sebastian Scherer" ], "categories": [ "cs.RO", "cs.CV" ], "pdf_url": "https://arxiv.org/pdf/2108.03936v1", "abstract": "Aerial vehicles are revolutionizing applications that require capturing the 3D structure of dynamic targets in the wild, such as sports, medicine, and entertainment. The core challenges in developing a motion-capture system that operates in outdoors environments are: (1) 3D inference requires multiple simultaneous viewpoints of the target, (2) occlusion caused by obstacles is frequent when tracking moving targets, and (3) the camera and vehicle state estimation is noisy. We present a real-time aerial system for multi-camera control that can reconstruct human motions in natural environments without the use of special-purpose markers. We develop a multi-robot coordination scheme that maintains the optimal flight formation for target reconstruction quality amongst obstacles. We provide studies evaluating system performance in simulation, and validate real-world performance using two drones while a target performs activities such as jogging and playing soccer. Supplementary video: https://youtu.be/jxt91vx0cns" }, "2107.08827.pdf": { "id": "2107.08827v1", "title": "Optimal sports betting strategies in practice: an experimental review", "year": "2021", "authors": [ "Matej Uhr\u00edn", "Gustav \u0160ourek", "Ond\u0159ej Hub\u00e1\u010dek", "Filip \u017delezn\u00fd" ], "categories": [ "q-fin.PM", "cs.CE", "q-fin.RM" ], "pdf_url": "https://arxiv.org/pdf/2107.08827v1", "abstract": "We investigate the most popular approaches to the problem of sports betting investment based on modern portfolio theory and the Kelly criterion. We define the problem setting, the formal investment strategies, and review their common modifications used in practice. The underlying purpose of the reviewed modifications is to mitigate the additional risk stemming from the unrealistic mathematical assumptions of the formal strategies. We test the resulting methods using a unified evaluation protocol for three sports: horse racing, basketball and soccer. The results show the practical necessity of the additional risk-control methods and demonstrate their individual benefits. Particularly, we show that an adaptive variant of the popular ``fractional Kelly'' method is a very suitable choice across a wide range of settings." }, "2106.04558.pdf": { "id": "2106.04558v1", "title": "Heatmaps in soccer: event vs tracking datasets", "year": "2021", "authors": [ "D. Garrido", "B. Burriel", "R. Resta", "R. Lopez del Campo", "J. M. Buldu" ], "categories": [ "physics.soc-ph", "physics.data-an" ], "pdf_url": "https://arxiv.org/pdf/2106.04558v1", "abstract": "We investigate how similar heatmaps of soccer players are when constructed from (i) event datasets and (ii) tracking datasets. When using event datasets, we show that the scale at which the events are grouped strongly influences the correlation with the tracking heatmaps. Furthermore, there is an optimal scale at which the correlation between event and tracking heatmaps is the highest. However, even at the optimal scale, correlations between both approaches are moderate. Furthermore, there is high heterogeneity in the players' correlation, ranging from negative values to correlations close to the unity. We show that the number of events performed by a player does not crucially determine the level of correlation between both heatmaps. Finally, we analyzed the influence of the player position, showing that defenders are the players with the highest correlations while forwards have the lowest." }, "2106.01972.pdf": { "id": "2106.01972v1", "title": "SOCCER: An Information-Sparse Discourse State Tracking Collection in the Sports Commentary Domain", "year": "2021", "authors": [ "Ruochen Zhang", "Carsten Eickhoff" ], "categories": [ "cs.CL" ], "pdf_url": "https://arxiv.org/pdf/2106.01972v1", "abstract": "In the pursuit of natural language understanding, there has been a long standing interest in tracking state changes throughout narratives. Impressive progress has been made in modeling the state of transaction-centric dialogues and procedural texts. However, this problem has been less intensively studied in the realm of general discourse where ground truth descriptions of states may be loosely defined and state changes are less densely distributed over utterances. This paper proposes to turn to simplified, fully observable systems that show some of these properties: Sports events. We curated 2,263 soccer matches including time-stamped natural language commentary accompanied by discrete events such as a team scoring goals, switching players or being penalized with cards. We propose a new task formulation where, given paragraphs of commentary of a game at different timestamps, the system is asked to recognize the occurrence of in-game events. This domain allows for rich descriptions of state while avoiding the complexities of many other real-world settings. As an initial point of performance measurement, we include two baseline methods from the perspectives of sentence classification with temporal dependence and current state-of-the-art generative model, respectively, and demonstrate that even sophisticated existing methods struggle on the state tracking task when the definition of state broadens or non-event chatter becomes prevalent." }, "2010.10739.pdf": { "id": "2010.10739v1", "title": "A Bayesian Hidden Semi-Markov Model with Covariate-Dependent State Duration Parameters for High-Frequency Data from Wearable Devices", "year": "2020", "authors": [ "Shirley Rojas-Salazar", "Erin M. Schliep", "Christopher K. Wikle", "Matthew Hawkey" ], "categories": [ "stat.AP" ], "pdf_url": "https://arxiv.org/pdf/2010.10739v1", "abstract": "Data collected by wearable devices in sports provide valuable information about an athlete's behavior such as their activity, performance, and ability. These time series data can be studied with approaches such as hidden Markov and semi-Markov models (HMM and HSMM) for varied purposes including activity recognition and event detection. HSMMs extend the HMM by explicitly modeling the time spent in each state. In a discrete-time HSMM, the duration in each state can be modeled with a zero-truncated Poisson distribution, where the duration parameter may be state-specific but constant in time. We extend the HSMM by allowing the state-specific duration parameters to vary in time and model them as a function of known covariates derived from the wearable device and observed over a period of time leading up to a state transition. In addition, we propose a data subsampling approach given that high-frequency data from wearable devices can violate the conditional independence assumption of the HSMM. We apply the model to wearable device data collected on a soccer referee in a Major League Soccer game. We model the referee's physiological response to the game demands and identify important time-varying effects of these demands associated with the duration in each state." }, "2009.13684.pdf": { "id": "2009.13684v2", "title": "Detecting soccer balls with reduced neural networks: a comparison of multiple architectures under constrained hardware scenarios", "year": "2020", "authors": [ "Douglas De Rizzo Meneghetti", "Thiago Pedro Donadon Homem", "Jonas Henrique Renolfi de Oliveira", "Isaac Jesus da Silva", "Danilo Hernani Perico", "Reinaldo Augusto da Costa Bianchi" ], "categories": [ "cs.CV" ], "pdf_url": "https://arxiv.org/pdf/2009.13684v2", "abstract": "Object detection techniques that achieve state-of-the-art detection accuracy employ convolutional neural networks, implemented to have optimal performance in graphics processing units. Some hardware systems, such as mobile robots, operate under constrained hardware situations, but still benefit from object detection capabilities. Multiple network models have been proposed, achieving comparable accuracy with reduced architectures and leaner operations. Motivated by the need to create an object detection system for a soccer team of mobile robots, this work provides a comparative study of recent proposals of neural networks targeted towards constrained hardware environments, in the specific task of soccer ball detection. We train multiple open implementations of MobileNetV2 and MobileNetV3 models with different underlying architectures, as well as YOLOv3, TinyYOLOv3, YOLOv4 and TinyYOLOv4 in an annotated image data set captured using a mobile robot. We then report their mean average precision on a test data set and their inference times in videos of different resolutions, under constrained and unconstrained hardware configurations. Results show that MobileNetV3 models have a good trade-off between mAP and inference time in constrained scenarios only, while MobileNetV2 with high width multipliers are appropriate for server-side inference. YOLO models in their official implementations are not suitable for inference in CPUs." }, "2009.06086.pdf": { "id": "2009.06086v1", "title": "Efficient Competitive Self-Play Policy Optimization", "year": "2020", "authors": [ "Yuanyi Zhong", "Yuan Zhou", "Jian Peng" ], "categories": [ "cs.LG", "cs.AI", "cs.GT", "stat.ML" ], "pdf_url": "https://arxiv.org/pdf/2009.06086v1", "abstract": "Reinforcement learning from self-play has recently reported many successes. Self-play, where the agents compete with themselves, is often used to generate training data for iterative policy improvement. In previous work, heuristic rules are designed to choose an opponent for the current learner. Typical rules include choosing the latest agent, the best agent, or a random historical agent. However, these rules may be inefficient in practice and sometimes do not guarantee convergence even in the simplest matrix games. In this paper, we propose a new algorithmic framework for competitive self-play reinforcement learning in two-player zero-sum games. We recognize the fact that the Nash equilibrium coincides with the saddle point of the stochastic payoff function, which motivates us to borrow ideas from classical saddle point optimization literature. Our method trains several agents simultaneously, and intelligently takes each other as opponent based on simple adversarial rules derived from a principled perturbation-based saddle optimization method. We prove theoretically that our algorithm converges to an approximate equilibrium with high probability in convex-concave games under standard assumptions. Beyond the theory, we further show the empirical superiority of our method over baseline methods relying on the aforementioned opponent-selection heuristics in matrix games, grid-world soccer, Gomoku, and simulated robot sumo, with neural net policy function approximators." }, "2009.02464.pdf": { "id": "2009.02464v1", "title": "PassVizor: Toward Better Understanding of the Dynamics of Soccer Passes", "year": "2020", "authors": [ "Xiao Xie", "Jiachen Wang", "Hongye Liang", "Dazhen Deng", "Shoubin Cheng", "Hui Zhang", "Wei Chen", "Yingcai Wu" ], "categories": [ "cs.HC" ], "pdf_url": "https://arxiv.org/pdf/2009.02464v1", "abstract": "In soccer, passing is the most frequent interaction between players and plays a significant role in creating scoring chances. Experts are interested in analyzing players' passing behavior to learn passing tactics, i.e., how players build up an attack with passing. Various approaches have been proposed to facilitate the analysis of passing tactics. However, the dynamic changes of a team's employed tactics over a match have not been comprehensively investigated. To address the problem, we closely collaborate with domain experts and characterize requirements to analyze the dynamic changes of a team's passing tactics. To characterize the passing tactic employed for each attack, we propose a topic-based approach that provides a high-level abstraction of complex passing behaviors. Based on the model, we propose a glyph-based design to reveal the multi-variate information of passing tactics within different phases of attacks, including player identity, spatial context, and formation. We further design and develop PassVizor, a visual analytics system, to support the comprehensive analysis of passing dynamics. With the system, users can detect the changing patterns of passing tactics and examine the detailed passing process for evaluating passing tactics. We invite experts to conduct analysis with PassVizor and demonstrate the usability of the system through an expert interview." }, "2007.12255.pdf": { "id": "2007.12255v1", "title": "Home Advantage in the Brazilian Elite Football: Verifying managers' capacity to outperform their disadvantage", "year": "2020", "authors": [ "Carlos Denner dos Santos", "Jessica Alves" ], "categories": [ "econ.GN" ], "pdf_url": "https://arxiv.org/pdf/2007.12255v1", "abstract": "Home advantage (HA) in football, soccer is well documented in the literature; however, the explanation for such phenomenon is yet to be completed, as this paper demonstrates that it is possible to overcome such disadvantage through managerial planning and intervention (tactics), an effect so far absent in the literature. To accomplish that, this study develops an integrative theoretical model of team performance to explain HA based on prior literature, pushing its limits to unfold the manager role in such a persistent pattern of performance in soccer. Data on one decade of the national tournament of Brazil was obtained from public sources, including information about matches and coaches of all 12 teams who played these competitions. Our conceptual modeling allows an empirical analysis of HA and performance that covers the effects of tactics, presence of supporters in matches and team fatigue via logistic regression. The results confirm the HA in the elite division of Brazilian soccer across all levels of comparative technical quality, a new variable introduced to control for a potential technical gap between teams (something that would turn the managerial influence null). Further analysis provides evidence that highlights managerial capacity to block the HA effect above and beyond the influences of fatigue (distance traveled) and density of people in the matches. This is the case of coaches Abel Braga, Marcelo Fernandes and Roger Machado, who were capable to reverse HA when playing against teams of similar quality. Overall, the home advantage diminishes as the comparative quality increases but disappears only when the two teams are of extremely different technical quality." }, "2005.04020.pdf": { "id": "2005.04020v2", "title": "Modeling ball possession dynamics in the game of football", "year": "2020", "authors": [ "A. Chacoma", "N. Almeira", "J. I. Perotti", "O. V. Billoni" ], "categories": [ "physics.soc-ph" ], "pdf_url": "https://arxiv.org/pdf/2005.04020v2", "abstract": "In this paper, we study interaction dynamics in the game of football-soccer in the context of ball possession intervals. To do so, we analyze a database comprising one season of the five major football leagues of Europe. Using this input, we developed a stochastic model based on three agents: two teammates and one defender. Despite its simplicity, the model is able to capture, in good approximation, the statistical behavior of possession times, pass lengths, and number of passes performed. In the last section, we show that the model's dynamics can be mapped into a Wiener process with drift and an absorbing barrier." }, "2004.09225.pdf": { "id": "2004.09225v5", "title": "Fairness in penalty shootouts: Is it worth using dynamic sequences?", "year": "2020", "authors": [ "L\u00e1szl\u00f3 Csat\u00f3", "D\u00f3ra Gr\u00e9ta Petr\u00f3czy" ], "categories": [ "cs.GT", "econ.GN" ], "pdf_url": "https://arxiv.org/pdf/2004.09225v5", "abstract": "The sequence of moves in a dynamic team tournament may distort the ex-ante winning probabilities and harm efficiency. This paper compares seven soccer penalty shootout rules that determine the kicking order, from a theoretical perspective. Their fairness is evaluated in a reasonable model of First Mover Advantage. We also discuss the probability of reaching the sudden death stage. In the case of stationary scoring probabilities, dynamic mechanisms are not better than static rules. However, it is worth compensating the second-mover by making it the first-mover in the sudden death stage. Our work has the potential to impact decision-makers who can guarantee fairer outcomes in dynamic tournaments by a carefully chosen sequence of actions." }, "2004.06172.pdf": { "id": "2004.06172v1", "title": "Event detection in coarsely annotated sports videos via parallel multi receptive field 1D convolutions", "year": "2020", "authors": [ "Kanav Vats", "Mehrnaz Fani", "Pascale Walters", "David A. Clausi", "John Zelek" ], "categories": [ "cs.CV" ], "pdf_url": "https://arxiv.org/pdf/2004.06172v1", "abstract": "In problems such as sports video analytics, it is difficult to obtain accurate frame level annotations and exact event duration because of the lengthy videos and sheer volume of video data. This issue is even more pronounced in fast-paced sports such as ice hockey. Obtaining annotations on a coarse scale can be much more practical and time efficient. We propose the task of event detection in coarsely annotated videos. We introduce a multi-tower temporal convolutional network architecture for the proposed task. The network, with the help of multiple receptive fields, processes information at various temporal scales to account for the uncertainty with regard to the exact event location and duration. We demonstrate the effectiveness of the multi-receptive field architecture through appropriate ablation studies. The method is evaluated on two tasks - event detection in coarsely annotated hockey videos in the NHL dataset and event spotting in soccer on the SoccerNet dataset. The two datasets lack frame-level annotations and have very distinct event frequencies. Experimental results demonstrate the effectiveness of the network by obtaining a 55% average F1 score on the NHL dataset and by achieving competitive performance compared to the state of the art on the SoccerNet dataset. We believe our approach will help develop more practical pipelines for event detection in sports video." }, "1910.10949.pdf": { "id": "1910.10949v1", "title": "ROBO: Robust, Fully Neural Object Detection for Robot Soccer", "year": "2019", "authors": [ "Marton Szemenyei", "Vladimir Estivill-Castro" ], "categories": [ "cs.CV", "cs.NE", "cs.RO" ], "pdf_url": "https://arxiv.org/pdf/1910.10949v1", "abstract": "Deep Learning has become exceptionally popular in the last few years due to its success in computer vision and other fields of AI. However, deep neural networks are computationally expensive, which limits their application in low power embedded systems, such as mobile robots. In this paper, an efficient neural network architecture is proposed for the problem of detecting relevant objects in robot soccer environments. The ROBO model's increase in efficiency is achieved by exploiting the peculiarities of the environment. Compared to the state-of-the-art Tiny YOLO model, the proposed network provides approximately 35 times decrease in run time, while achieving superior average precision, although at the cost of slightly worse localization accuracy." }, "1910.10232.pdf": { "id": "1910.10232v2", "title": "Bottom-Up Meta-Policy Search", "year": "2019", "authors": [ "Luckeciano C. Melo", "Marcos R. O. A. Maximo", "Adilson Marques da Cunha" ], "categories": [ "cs.LG", "cs.AI", "cs.RO" ], "pdf_url": "https://arxiv.org/pdf/1910.10232v2", "abstract": "Despite of the recent progress in agents that learn through interaction, there are several challenges in terms of sample efficiency and generalization across unseen behaviors during training. To mitigate these problems, we propose and apply a first-order Meta-Learning algorithm called Bottom-Up Meta-Policy Search (BUMPS), which works with two-phase optimization procedure: firstly, in a meta-training phase, it distills few expert policies to create a meta-policy capable of generalizing knowledge to unseen tasks during training; secondly, it applies a fast adaptation strategy named Policy Filtering, which evaluates few policies sampled from the meta-policy distribution and selects which best solves the task. We conducted all experiments in the RoboCup 3D Soccer Simulation domain, in the context of kick motion learning. We show that, given our experimental setup, BUMPS works in scenarios where simple multi-task Reinforcement Learning does not. Finally, we performed experiments in a way to evaluate each component of the algorithm." }, "1910.07834.pdf": { "id": "1910.07834v1", "title": "Using a KG-Copy Network for Non-Goal Oriented Dialogues", "year": "2019", "authors": [ "Debanjan Chaudhuri", "Md Rashad Al Hasan Rony", "Simon Jordan", "Jens Lehmann" ], "categories": [ "cs.CL" ], "pdf_url": "https://arxiv.org/pdf/1910.07834v1", "abstract": "Non-goal oriented, generative dialogue systems lack the ability to generate answers with grounded facts. A knowledge graph can be considered an abstraction of the real world consisting of well-grounded facts. This paper addresses the problem of generating well grounded responses by integrating knowledge graphs into the dialogue systems response generation process, in an end-to-end manner. A dataset for nongoal oriented dialogues is proposed in this paper in the domain of soccer, conversing on different clubs and national teams along with a knowledge graph for each of these teams. A novel neural network architecture is also proposed as a baseline on this dataset, which can integrate knowledge graphs into the response generation process, producing well articulated, knowledge grounded responses. Empirical evidence suggests that the proposed model performs better than other state-of-the-art models for knowledge graph integrated dialogue systems." }, "1909.02406.pdf": { "id": "1909.02406v2", "title": "Utilizing Temporal Information in Deep Convolutional Network for Efficient Soccer Ball Detection and Tracking", "year": "2019", "authors": [ "Anna Kukleva", "Mohammad Asif Khan", "Hafez Farazi", "Sven Behnke" ], "categories": [ "cs.CV" ], "pdf_url": "https://arxiv.org/pdf/1909.02406v2", "abstract": "Soccer ball detection is identified as one of the critical challenges in the RoboCup competition. It requires an efficient vision system capable of handling the task of detection with high precision and recall and providing robust and low inference time. In this work, we present a novel convolutional neural network (CNN) approach to detect the soccer ball in an image sequence. In contrast to the existing methods where only the current frame or an image is used for the detection, we make use of the history of frames. Using history allows to efficiently track the ball in situations where the ball disappears or gets partially occluded in some of the frames. Our approach exploits spatio-temporal correlation and detects the ball based on the trajectory of its movements. We present our results with three convolutional methods, namely temporal convolutional networks (TCN), ConvLSTM, and ConvGRU. We first solve the detection task for an image using fully convolutional encoder-decoder architecture, and later, we use it as an input to our temporal models and jointly learn the detection task in sequences of images. We evaluate all our experiments on a novel dataset prepared as a part of this work. Furthermore, we present empirical results to support the effectiveness of using the history of the ball in challenging scenarios." }, "1909.01788.pdf": { "id": "1909.01788v2", "title": "Fractals2019: Combinatorial Optimisation with Dynamic Constraint Annealing", "year": "2019", "authors": [ "Mikhail Prokopenko", "Peter Wang" ], "categories": [ "cs.AI", "cs.DM", "cs.MA" ], "pdf_url": "https://arxiv.org/pdf/1909.01788v2", "abstract": "Fractals2019 started as a new experimental entry in the RoboCup Soccer 2D Simulation League, based on Gliders2d code base, and advanced to become a RoboCup-2019 champion. We employ combinatorial optimisation methods, within the framework of Guided Self-Organisation, with the search guided by local constraints. We present examples of several tactical tasks based on the Gliders2d code (version v2), including the search for an optimal assignment of heterogeneous player types, as well as blocking behaviours, offside trap, and attacking formations. We propose a new method, Dynamic Constraint Annealing, for solving dynamic constraint satisfaction problems, and apply it to optimise thermodynamic potential of collective behaviours, under dynamically induced constraints." }, "1907.09333.pdf": { "id": "1907.09333v1", "title": "Continuously Updated Data Analysis Systems", "year": "2019", "authors": [ "Lee F. Richardson" ], "categories": [ "stat.OT", "stat.CO" ], "pdf_url": "https://arxiv.org/pdf/1907.09333v1", "abstract": "When doing data science, it's important to know what you're building. This paper describes an idealized final product of a data science project, called a Continuously Updated Data-Analysis System (CUDAS). The CUDAS concept synthesizes ideas from a range of successful data science projects, such as Nate Silver's FiveThirtyEight. A CUDAS can be built for any context, such as the state of the economy, the state of the climate, and so on. To demonstrate, we build two CUDAS systems. The first provides continuously-updated ratings for soccer players, based on the newly developed Augmented Adjusted Plus-Minus statistic. The second creates a large dataset of synthetic ecosystems, which is used for agent-based modeling of infectious diseases." }, "1907.00327.pdf": { "id": "1907.00327v1", "title": "Collaboration of AI Agents via Cooperative Multi-Agent Deep Reinforcement Learning", "year": "2019", "authors": [ "Niranjan Balachandar", "Justin Dieter", "Govardana Sachithanandam Ramachandran" ], "categories": [ "cs.MA", "cs.AI", "cs.CV" ], "pdf_url": "https://arxiv.org/pdf/1907.00327v1", "abstract": "There are many AI tasks involving multiple interacting agents where agents should learn to cooperate and collaborate to effectively perform the task. Here we develop and evaluate various multi-agent protocols to train agents to collaborate with teammates in grid soccer. We train and evaluate our multi-agent methods against a team operating with a smart hand-coded policy. As a baseline, we train agents concurrently and independently, with no communication. Our collaborative protocols were parameter sharing, coordinated learning with communication, and counterfactual policy gradients. Against the hand-coded team, the team trained with parameter sharing and the team trained with coordinated learning performed the best, scoring on 89.5% and 94.5% of episodes respectively when playing against the hand-coded team. Against the parameter sharing team, with adversarial training the coordinated learning team scored on 75% of the episodes, indicating it is the most adaptable of our methods. The insights gained from our work can be applied to other domains where multi-agent collaboration could be beneficial." }, "1905.09934.pdf": { "id": "1905.09934v1", "title": "Mechatronic Design of a Dribbling System for RoboCup Small Size Robot", "year": "2019", "authors": [ "Zheyuan Huang", "Yunkai Wang", "Lingyun Chen", "Jiacheng Li", "Zexi Chen", "Rong Xiong" ], "categories": [ "cs.RO" ], "pdf_url": "https://arxiv.org/pdf/1905.09934v1", "abstract": "RoboCup SSL is an excellent platform for researching artificial intelligence and robotics. The dribbling system is an essential issue, which is the main part for completing advanced soccer skills such as trapping and dribbling. In this paper, we designed a new dribbling system for SSL robots, including mechatronics design and control algorithms. For the mechatronics design, we analysed and exposed the 3-touch-point model with the simulation in ADAMS. In the motor controller algorithm, we use reinforcement learning to control the torque output. Finally we verified the results on the robot." }, "1902.07304.pdf": { "id": "1902.07304v1", "title": "DeepBall: Deep Neural-Network Ball Detector", "year": "2019", "authors": [ "Jacek Komorowski", "Grzegorz Kurzejamski", "Grzegorz Sarwas" ], "categories": [ "cs.CV" ], "pdf_url": "https://arxiv.org/pdf/1902.07304v1", "abstract": "The paper describes a deep network based object detector specialized for ball detection in long shot videos. Due to its fully convolutional design, the method operates on images of any size and produces \\emph{ball confidence map} encoding the position of detected ball. The network uses hypercolumn concept, where feature maps from different hierarchy levels of the deep convolutional network are combined and jointly fed to the convolutional classification layer. This allows boosting the detection accuracy as larger visual context around the object of interest is taken into account. The method achieves state-of-the-art results when tested on publicly available ISSIA-CNR Soccer Dataset." }, "1902.07151.pdf": { "id": "1902.07151v2", "title": "Emergent Coordination Through Competition", "year": "2019", "authors": [ "Siqi Liu", "Guy Lever", "Josh Merel", "Saran Tunyasuvunakool", "Nicolas Heess", "Thore Graepel" ], "categories": [ "cs.AI" ], "pdf_url": "https://arxiv.org/pdf/1902.07151v2", "abstract": "We study the emergence of cooperative behaviors in reinforcement learning agents by introducing a challenging competitive multi-agent soccer environment with continuous simulated physics. We demonstrate that decentralized, population-based training with co-play can lead to a progression in agents' behaviors: from random, to simple ball chasing, and finally showing evidence of cooperation. Our study highlights several of the challenges encountered in large scale multi-agent training in continuous control. In particular, we demonstrate that the automatic optimization of simple shaping rewards, not themselves conducive to co-operative behavior, can lead to long-horizon team behavior. We further apply an evaluation scheme, grounded by game theoretic principals, that can assess agent performance in the absence of pre-defined evaluation tasks or human baselines." }, "1807.00381.pdf": { "id": "1807.00381v1", "title": "Model-based Exception Mining for Object-Relational Data", "year": "2018", "authors": [ "Fatemeh Riahi", "Oliver Schulte" ], "categories": [ "cs.AI", "cs.LG", "stat.ML" ], "pdf_url": "https://arxiv.org/pdf/1807.00381v1", "abstract": "This paper is based on a previous publication [29]. Our work extends exception mining and outlier detection to the case of object-relational data. Object-relational data represent a complex heterogeneous network [12], which comprises objects of different types, links among these objects, also of different types, and attributes of these links. This special structure prohibits a direct vectorial data representation. We follow the well-established Exceptional Model Mining framework, which leverages machine learning models for exception mining: A object is exceptional to the extent that a model learned for the object data differs from a model learned for the general population. Exceptional objects can be viewed as outliers. We apply state of-the-art probabilistic modelling techniques for object-relational data that construct a graphical model (Bayesian network), which compactly represents probabilistic associations in the data. A new metric, derived from the learned object-relational model, quantifies the extent to which the individual association pattern of a potential outlier deviates from that of the whole population. The metric is based on the likelihood ratio of two parameter vectors: One that represents the population associations, and another that represents the individual associations. Our method is validated on synthetic datasets and on real-world data sets about soccer matches and movies. Compared to baseline methods, our novel transformed likelihood ratio achieved the best detection accuracy on all datasets." }, "1805.09910.pdf": { "id": "1805.09910v1", "title": "Fairness GAN", "year": "2018", "authors": [ "Prasanna Sattigeri", "Samuel C. Hoffman", "Vijil Chenthamarakshan", "Kush R. Varshney" ], "categories": [ "stat.ML", "cs.CY", "cs.LG" ], "pdf_url": "https://arxiv.org/pdf/1805.09910v1", "abstract": "In this paper, we introduce the Fairness GAN, an approach for generating a dataset that is plausibly similar to a given multimedia dataset, but is more fair with respect to protected attributes in allocative decision making. We propose a novel auxiliary classifier GAN that strives for demographic parity or equality of opportunity and show empirical results on several datasets, including the CelebFaces Attributes (CelebA) dataset, the Quick, Draw!\\ dataset, and a dataset of soccer player images and the offenses they were called for. The proposed formulation is well-suited to absorbing unlabeled data; we leverage this to augment the soccer dataset with the much larger CelebA dataset. The methodology tends to improve demographic parity and equality of opportunity while generating plausible images." }, "1803.01203.pdf": { "id": "1803.01203v1", "title": "Multiresolution Tensor Decomposition for Multiple Spatial Passing Networks", "year": "2018", "authors": [ "Shaobo Han", "David B. Dunson" ], "categories": [ "stat.AP", "cs.LG", "cs.SI", "stat.ML" ], "pdf_url": "https://arxiv.org/pdf/1803.01203v1", "abstract": "This article is motivated by soccer positional passing networks collected across multiple games. We refer to these data as replicated spatial passing networks---to accurately model such data it is necessary to take into account the spatial positions of the passer and receiver for each passing event. This spatial registration and replicates that occur across games represent key differences with usual social network data. As a key step before investigating how the passing dynamics influence team performance, we focus on developing methods for summarizing different team's passing strategies. Our proposed approach relies on a novel multiresolution data representation framework and Poisson nonnegative block term decomposition model, which automatically produces coarse-to-fine low-rank network motifs. The proposed methods are applied to detailed passing record data collected from the 2014 FIFA World Cup." }, "1704.00583.pdf": { "id": "1704.00583v1", "title": "A PageRank Model for Player Performance Assessment in Basketball, Soccer and Hockey", "year": "2017", "authors": [ "Shael Brown" ], "categories": [ "stat.AP" ], "pdf_url": "https://arxiv.org/pdf/1704.00583v1", "abstract": "In the sports of soccer, hockey and basketball the most commonly used statistics for player performance assessment are divided into two categories: offensive statistics and defensive statistics. However, qualitative assessments of playmaking (for example making \"smart\" passes) are difficult to quantify. It would be advantageous to have available a single statistic that can emphasize the flow of a game, rewarding those players who initiate and contribute to successful plays more. In this paper we will examine a model based on Google's PageRank. Other papers have explored ranking teams, coaches, and captains but here we construct ratings and rankings for individual members on both teams that emphasizes initiating and partaking in successful plays and forcing defensive turnovers. For a soccer/hockey/basketball game, our model assigns a node for each of the n players who play in the game and a \"goal node\". Arcs between player nodes indicate sport-specific situations (including passes, turnovers, scoring, fouls, out-of-bounds, play-stoppages, turnovers, missed shots, defensive plays etc.), tailored for each sport. As well, some additional arcs are added in to ensure that the associated matrix is primitive and hence there is a unique PageRank vector. The PageRank vector of the associated matrix is used to rank the players of the game. To illustrate the model, data was taken from nine NBA games played between 2014-2016. Many of the top-ranked players (in the model) in a given game had some of the most impressive traditional stat-lines. However, from the model there were surprises where some players who had impressive stat-lines had lower ranks, and others who had less impressive stat-lines had higher ranks. Overall, the model's ranking and ratings reflect more the flow of the game compared to traditional sports statistics." }, "1408.4225.pdf": { "id": "1408.4225v1", "title": "Analytic Model of Variable Characteristic of Coefficient of Restitution and Its Application to Soccer Ball Trajectory Planning", "year": "2014", "authors": [ "Hwan-Taek Ryu", "Byung-Ju Yi", "Younghun Kwon" ], "categories": [ "physics.class-ph" ], "pdf_url": "https://arxiv.org/pdf/1408.4225v1", "abstract": "In this article, we investigate the behavior of the coefficient of restitution (COR) which is an important parameter in many impact-related fields. In many cases, the COR is considered as a constant value, but it varies according to many variables. In this paper, we introduce an analytical variable COR model considering aero dynamics along with its verification through experiment. To introduce and analyze the variable characteristic of the COR model, the collision phenomenon between a pendulum and two kinds of ball is employed as an example and aerodynamics such as drag force is considered for analyzing the after-effect of the collision. Collision velocity of the pendulum, dynamic parameters of colliding bodies, contact time, drag coefficient, the air density, and the cross-sectional area of the ball are found as the typical variables of analytical COR model. This observation generalizes the result in previous researches. To verify new COR model, the travel distances for the curve-fitted constant COR model and the curve-fitted variable COR model are compared through simulation and experiment. Moreover, comparison between constant COR and variable COR is presented in several point of views. Finally, using the variable COR model, the travel distance of the ball for collision velocity, which is beyond the curve-fitted range, is estimated." }, "1406.4644.pdf": { "id": "1406.4644v1", "title": "Primacy & Ranking of UEFA Soccer Teams from Biasing Organizing Rules", "year": "2014", "authors": [ "Marcel Ausloos", "Adam Gadomski", "Nikolay K. Vitanov" ], "categories": [ "physics.soc-ph", "physics.data-an" ], "pdf_url": "https://arxiv.org/pdf/1406.4644v1", "abstract": "A question is raised on whether some implied regularity or structure, as found in soccer team ranking by the Union of European Football Associations (UEFA), is due to implicit game result value or score competition conditions. The analysis is based on considerations about complex systems, i.e. searching whether power or other simple law fits are appropriate to describe some internal dynamics. It is observed that the ranking is specifically organized: a major class made of a few teams emerges after each game season. Other classes which apparently have regular sizes subsequently occur. Thus, the notion of Sheppard primacy index is envisaged to describe the findings. Additional primacy indices are discussed for enhancing the features. These measures can be used to sort out peer classes in more general terms. A very simplified toy model containing ingredients of the UEFA ranking rules suggests that such peer classes are an extrinsic property of the ranking, as obtained in many nonlinear systems under boundary condition constraints." }, "1307.3412.pdf": { "id": "1307.3412v2", "title": "A new method for comparing rankings through complex networks: Model and analysis of competitiveness of major European soccer leagues", "year": "2013", "authors": [ "Regino Criado", "Esther Garcia", "Francisco Pedroche", "Miguel Romance" ], "categories": [ "physics.soc-ph", "cs.DM", "cs.SI" ], "pdf_url": "https://arxiv.org/pdf/1307.3412v2", "abstract": "In this paper we show a new technique to analyze families of rankings. In particular we focus on sports rankings and, more precisely, on soccer leagues. We consider that two teams compete when they change their relative positions in consecutive rankings. This allows to define a graph by linking teams that compete. We show how to use some structural properties of this competitivity graph to measure to what extend the teams in a league compete. These structural properties are the mean degree, the mean strength and the clustering coefficient. We give a generalization of the Kendall's correlation coefficient to more than two rankings. We also show how to make a dynamic analysis of a league and how to compare different leagues. We apply this technique to analyze the four major European soccer leagues: Bundesliga, Italian Lega, Spanish Liga, and Premier League. We compare our results with the classical analysis of sport ranking based on measures of competitive balance." }, "1207.6796.pdf": { "id": "1207.6796v1", "title": "Around the goal: Examining the effect of the first goal on the second goal in soccer using survival analysis methods", "year": "2012", "authors": [ "Daniel Nevo", "Ya'acov Ritov" ], "categories": [ "stat.AP" ], "pdf_url": "https://arxiv.org/pdf/1207.6796v1", "abstract": "In this paper we apply survival techniques to soccer data, treating a goal scoring as the event of interest. It specifically concerns the relationship between the time of the first goal in the game and the time of the second goal. In order to do so, the relevant survival analysis concepts are readjusted to fit the problem and a Cox model is developed for the hazard function. Attributes such as time dependent covariates and a frailty term are also being considered. We also use a reliable propensity score to summarize the pre-game covariates. The conclusions derived from the results are that a first goal occurrence could either expedite or impede the next goal scoring, depending on the time it was scored. Moreover, once a goal is scored, another goal scoring become more and more likely as the game progresses. Furthermore, the first goal effect is the same whether the goal was scored or conceded." }, "1207.1848.pdf": { "id": "1207.1848v1", "title": "A Simple Non-Markovian Computational Model of the Statistics of Soccer Leagues: Emergence and Scaling effects", "year": "2012", "authors": [ "Roberto da Silva", "Mendeli Vainstein", "Luis Lamb", "Sandra Prado" ], "categories": [ "physics.data-an", "cond-mat.stat-mech", "physics.soc-ph" ], "pdf_url": "https://arxiv.org/pdf/1207.1848v1", "abstract": "We propose a novel algorithm that outputs the final standings of a soccer league, based on a simple dynamics that mimics a soccer tournament. In our model, a team is created with a defined potential(ability) which is updated during the tournament according to the results of previous games. The updated potential modifies a teams' future winning/losing probabilities. We show that this evolutionary game is able to reproduce the statistical properties of final standings of actual editions of the Brazilian tournament (Brasileir\u00e3o). However, other leagues such as the Italian and the Spanish tournaments have notoriously non-Gaussian traces and cannot be straightforwardly reproduced by this evolutionary non-Markovian model. A complete understanding of these phenomena deserves much more attention, but we suggest a simple explanation based on data collected in Brazil: Here several teams were crowned champion in previous editions corroborating that the champion typically emerges from random fluctuations that partly preserves the gaussian traces during the tournament. On the other hand, in the Italian and Spanish leagues only a few teams in recent history have won their league tournaments. These leagues are based on more robust and hierarchical structures established even before the beginning of the tournament. For the sake of completeness, we also elaborate a totally Gaussian model (which equalizes the winning, drawing, and losing probabilities) and we show that the scores of the \"Brasileir\u00e3o\" cannot be reproduced. Such aspects stress that evolutionary aspects are not superfluous in our modeling. Finally, we analyse the distortions of our model in situations where a large number of teams is considered, showing the existence of a transition from a single to a double peaked histogram of the final classification scores. An interesting scaling is presented for different sized tournaments." }, "1004.2003.pdf": { "id": "1004.2003v2", "title": "The Socceral Force", "year": "2010", "authors": [ "Norbert B\u00e1tfai" ], "categories": [ "cs.AI", "cs.SE" ], "pdf_url": "https://arxiv.org/pdf/1004.2003v2", "abstract": "We have an audacious dream, we would like to develop a simulation and virtual reality system to support the decision making in European football (soccer). In this review, we summarize the efforts that we have made to fulfil this dream until recently. In addition, an introductory version of FerSML (Footballer and Football Simulation Markup Language) is presented in this paper." }, "2609.15541.pdf": { "id": "2609.15541v1", "title": "The life and death of football team runs: survival of collective modes shapes L\u00e9vy-like transport", "year": "2026", "authors": [ "Guillermo Herrera S\u00e1nchez", "Daniel Gradeci", "Daniele Ramsay", "Safe Khan" ], "categories": [ "physics.soc-ph" ], "pdf_url": "https://arxiv.org/pdf/2609.15541v1", "abstract": "Broad run-length distributions and short-lag superdiffusion have recently been reported in football players and team centroids, prompting a collective-foraging interpretation. The mechanism linking these player- and team-level signatures remains unclear. Using SoccerMon GPS data from 66 tracked team-match records across 62 fixtures in the Norwegian women's top-flight Toppserien, we asked whether player transport is inherited from a translating team mode and what controls the lifetime of that mode. We decomposed player displacement into centroid translation and motion within the formation, then modelled the switching and termination of directional centroid runs. At longer lags, centroid translation carried an increasing share of player displacement. Run lifetimes were broad but finite: termination was highest near onset, declined with age and rose modestly later. After accounting for speed and directional persistence, collective order added only limited predictive information. At the population level, an age-dependent switching-and-termination model predicted, on held-out match dates, the early enrichment and later depletion of high-order states among surviving runs. These results link player- and team-level L\u00e9vy-like movement through a finite-lived collective mode whose age and internal state shape its survival. Long displacements need not be selected in advance; they can emerge when transient collective modes persist, reorganise and selectively survive. More broadly, trajectory statistics may record which dynamical histories survive, not only the rules by which agents move." }, "2108.05575.pdf": { "id": "2108.05575v1", "title": "Kicktionary-LOME: A Domain-Specific Multilingual Frame Semantic Parsing Model for Football Language", "year": "2021", "authors": [ "Gosse Minnema" ], "categories": [ "cs.CL" ], "pdf_url": "https://arxiv.org/pdf/2108.05575v1", "abstract": "This technical report introduces an adapted version of the LOME frame semantic parsing model (Xia et al., EACL 2021) which is capable of automatically annotating texts according to the \"Kicktionary\" domain-specific framenet resource. Several methods for training a model even with limited available training data are proposed. While there are some challenges for evaluation related to the nature of the available annotations, preliminary results are very promising, with the best model reaching F1-scores of 0.83 (frame prediction) and 0.81 (semantic role prediction)." }, "1805.08937.pdf": { "id": "1805.08937v1", "title": "Predicting football tables by a maximally parsimonious model", "year": "2018", "authors": [ "Kjetil Haugen", "Brynjulf Owren" ], "categories": [ "stat.AP" ], "pdf_url": "https://arxiv.org/pdf/1805.08937v1", "abstract": "This paper presents some useful mathematical results involved in football table prediction. In addition, some empirical results indicate that an alternative methodology for football table prediction may produce high quality forecasts with far less resource usage than conventional methods." }, "1710.00001.pdf": { "id": "1710.00001v2", "title": "A Bayesian inference approach for determining player abilities in football", "year": "2017", "authors": [ "Gavin A. Whitaker", "Ricardo Silva", "Daniel Edwards", "Ioannis Kosmidis" ], "categories": [ "stat.AP" ], "pdf_url": "https://arxiv.org/pdf/1710.00001v2", "abstract": "We consider the task of determining a football player's ability for a given event type, for example, scoring a goal. We propose an interpretable Bayesian model which is fit using variational inference methods. We implement a Poisson model to capture occurrences of event types, from which we infer player abilities. Our approach also allows the visualisation of differences between players, for a specific ability, through the marginal posterior variational densities. We then use these inferred player abilities to extend the Bayesian hierarchical model of Baio and Blangiardo (2010) which captures a team's scoring rate (the rate at which they score goals). We apply the resulting scheme to the English Premier League, capturing player abilities over the 2013/2014 season, before using output from the hierarchical model to predict whether over or under 2.5 goals will be scored in a given game in the 2014/2015 season. This validates our model as a way of providing insights into team formation and the individual success of sports teams." }, "2211.06052.pdf": { "id": "2211.06052v1", "title": "Gambling on Momentum", "year": "2022", "authors": [ "Marius \u00d6tting", "Christian Deutscher", "Carl Singleton", "Luca De Angelis" ], "categories": [ "econ.GN" ], "pdf_url": "https://arxiv.org/pdf/2211.06052v1", "abstract": "Sports betting markets are proven real-world laboratories to test theories of asset pricing anomalies and risky behaviour. Using a high-frequency dataset provided directly by a major bookmaker, containing the odds and amounts staked throughout German Bundesliga football matches, we test for evidence of momentum in the betting and pricing behaviour after equalising goals. We find that bettors see value in teams that have the apparent momentum, staking about 40% more on them than teams that just conceded an equaliser. Still, there is no evidence that such perceived momentum matters on average for match outcomes or is associated with the bookmaker offering favourable odds. We also confirm that betting on the apparent momentum would lead to substantial losses for bettors." }, "2108.00821.pdf": { "id": "2108.00821v2", "title": "The reaction to news in live betting", "year": "2021", "authors": [ "Marius \u00d6tting", "Rouven Michels", "Roland Langrock", "Christian Deutscher" ], "categories": [ "physics.soc-ph" ], "pdf_url": "https://arxiv.org/pdf/2108.00821v2", "abstract": "Sports betting markets have grown very rapidly recently, with the total European gambling market worth 98.6 billion euro in 2019. Considering a high-resolution (1 Hz) data set provided by a large European bookmaker, we investigate the effect of news on the dynamics of live betting. In particular, we consider stakes placed in a live betting market during football matches. Accounting for the general market activity level within a state-space modelling framework, we focus on the market's response to events such as goals (i.e. major news), but also to the general situation within a match such as the uncertainty about the outcome. Our results indicate that markets might overreact to recent news, confirming cognitive biases known from psychology and behavioural economics." }, "1701.07555.pdf": { "id": "1701.07555v1", "title": "Robust analysis of second-leg home advantage in UEFA football through better nonparametric confidence intervals for binary regression functions", "year": "2017", "authors": [ "Gery Geenens", "Thomas Cuddihy" ], "categories": [ "stat.ME", "stat.AP" ], "pdf_url": "https://arxiv.org/pdf/1701.07555v1", "abstract": "In international football (soccer), two-legged knockout ties, with each team playing at home in one leg and the final outcome decided on aggregate, are common. Many players, managers and followers seem to believe in the `second-leg home advantage', i.e. that it is beneficial to play at home on the second leg. A more complex effect than the usual and well-established home advantage, it is harder to identify, and previous statistical studies did not prove conclusive about its actuality. Yet, given the amount of money handled in international football competitions nowadays, the question of existence or otherwise of this effect is of real import. As opposed to previous research, this paper addresses it from a purely nonparametric perspective and brings a very objective answer, not based on any particular model specification which could orientate the analysis in one or the other direction. Along the way, the paper reviews the well-known shortcomings of the Wald confidence interval for a proportion, suggests new nonparametric confidence intervals for conditional probability functions, revisits the problem of the bias when building confidence intervals in nonparametric regression, and provides a novel bootstrap-based solution to it. Finally, the new intervals are used in a careful analysis of game outcome data for the UEFA Champions and Europa leagues from 2009/10 to 2014/15. A slight `second-leg home advantage' is evidenced." }, "1305.4955.pdf": { "id": "1305.4955v2", "title": "A Data Mining Approach to Solve the Goal Scoring Problem", "year": "2013", "authors": [ "Renato Oliveira", "Paulo Adeodato", "Arthur Carvalho", "Icamaan Viegas", "Christian Diego", "Tsang Ing-Ren" ], "categories": [ "cs.AI", "cs.LG" ], "pdf_url": "https://arxiv.org/pdf/1305.4955v2", "abstract": "In soccer, scoring goals is a fundamental objective which depends on many conditions and constraints. Considering the RoboCup soccer 2D-simulator, this paper presents a data mining-based decision system to identify the best time and direction to kick the ball towards the goal to maximize the overall chances of scoring during a simulated soccer match. Following the CRISP-DM methodology, data for modeling were extracted from matches of major international tournaments (10691 kicks), knowledge about soccer was embedded via transformation of variables and a Multilayer Perceptron was used to estimate the scoring chance. Experimental performance assessment to compare this approach against previous LDA-based approach was conducted from 100 matches. Several statistical metrics were used to analyze the performance of the system and the results showed an increase of 7.7% in the number of kicks, producing an overall increase of 78% in the number of goals scored." }, "2607.01722.pdf": { "id": "2607.01722v2", "title": "An Adaptive Glicko-2 Rating Framework for Probabilistic Football Forecasting and Season Simulation", "year": "2026", "authors": [ "Bich Van Nguyen", "Nam Anh Tran" ], "categories": [ "stat.AP" ], "pdf_url": "https://arxiv.org/pdf/2607.01722v2", "abstract": "Football match outcome prediction is a challenging problem because team strength changes over time, match outcomes contain a high level of randomness, and draws play a central role in the result structure. Classical rating systems such as Elo provide simple and interpretable dynamic summaries of team ability, but they do not explicitly model uncertainty and often ignore football-specific contextual information. This paper proposes an adaptive Glicko-2-based rating framework for probabilistic football forecasting and leaguelevel season simulation. The proposed framework extends the standard Glicko-2 model by incorporating football-specific mechanisms, including margin-of-victory adjustment, dominance weighting, structural shocks, home advantage modelling, and an ordered-logit draw model. The framework estimates latent team strength dynamically, converts rating differences into win-draw-loss probabilities, and uses these probabilities to simulate the remaining part of a league season through Monte Carlo sampling." } }