Data-driven Analysis for Understanding Team Sports Behaviors

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.

arXiv comment9 pages, 2 figures. This is the first draft and the final version will be published in the Journal of Robotics and Mechatronics

References in corpus (24)

• Social Force Model for Pedestrian Dynamics
• Human Motion Trajectory Prediction: A Survey
• Spatio-Temporal Analysis of Team Sports -- A Survey
• Actions Speak Louder Than Goals: Valuing Player Actions in Soccer
• PlayeRank: data-driven performance evaluation and player ranking in soccer via a machine learning approach
• A Multiresolution Stochastic Process Model for Predicting Basketball Possession Outcomes
• A Disentangled Recognition and Nonlinear Dynamics Model for Unsupervised Learning
• Learning Koopman Invariant Subspaces for Dynamic Mode Decomposition
• Generating Long-term Trajectories Using Deep Hierarchical Networks
• Deep Reinforcement Learning in Ice Hockey for Context-Aware Player Evaluation
• Generating Multi-Agent Trajectories using Programmatic Weak Supervision
• Hierarchical Deep Multiagent Reinforcement Learning with Temporal Abstraction
• Dynamic mode decomposition in vector-valued reproducing kernel Hilbert spaces for extracting dynamical structure among observables
• VAIN: Attentional Multi-agent Predictive Modeling
• Emergence of self-similarity in football dynamics
• Cracking the Black Box: Distilling Deep Sports Analytics
• Physically-interpretable classification of biological network dynamics for complex collective motions
• NAOMI: Non-Autoregressive Multiresolution Sequence Imputation
• Fine-Grained Retrieval of Sports Plays using Tree-Based Alignment of Trajectories
• The Advantage of Doubling: A Deep Reinforcement Learning Approach to Studying the Double Team in the NBA
• Multiresolution Tensor Learning for Efficient and Interpretable Spatial Analysis
• baller2vec: A Multi-Entity Transformer For Multi-Agent Spatiotemporal Modeling
• Improved Structural Discovery and Representation Learning of Multi-Agent Data
• Multi-resolution Tensor Learning for Large-Scale Spatial Data

Cited by in corpus (3)

• Action valuation of on- and off-ball soccer players based on multi-agent deep reinforcement learning
• Decentralized policy learning with partial observation and mechanical constraints for multiperson modeling
• PassAI: explainable artificial intelligence algorithm for soccer pass analysis using multimodal information resources