Optimising Long-Term Outcomes using Real-World Fluent Objectives: An Application to Football

In this paper, we present a novel approach for optimising longterm 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%. ACM Reference Format: Ryan Beal, Georgios Chalkiadakis, Timothy J. Norman, and Sarvapali D. Ramchurn. 2021. Optimising Long-Term Outcomes using Real-World Fluent Objectives: An Application to Football. In Proc. of the 20th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2021), Online, May 3–7, 2021, IFAAMAS, 9 pages.

[1]  Sarvapali D. Ramchurn,et al.  Competing with Humans at Fantasy Football: Team Formation in Large Partially-Observable Domains , 2012, AAAI.

[2]  D. JordanJeremy,et al.  Optimizing Football Game Play Calling , 2009 .

[3]  Manish Jain,et al.  Computing optimal randomized resource allocations for massive security games , 2009, AAMAS.

[4]  Kwang Mong Sim,et al.  Agents that react to changing market situations , 2003, IEEE Trans. Syst. Man Cybern. Part B.

[5]  Matthieu Geist,et al.  Learning from Demonstrations: Is It Worth Estimating a Reward Function? , 2013, ECML/PKDD.

[6]  Chin-Hui Lee,et al.  Maximum a posteriori estimation for multivariate Gaussian mixture observations of Markov chains , 1994, IEEE Trans. Speech Audio Process..

[7]  Herbert Mayo,et al.  Risk-Adjusted Returns and Stock Market Games , 1995 .

[8]  D. Higdon,et al.  Accelerating Markov Chain Monte Carlo Simulation by Differential Evolution with Self-Adaptive Randomized Subspace Sampling , 2009 .

[9]  Javier Fernández,et al.  Decomposing the Immeasurable Sport: A deep learning expected possession value framework for soccer , 2019 .

[10]  M. Marchesi,et al.  Scaling and criticality in a stochastic multi-agent model of a financial market , 1999, Nature.

[11]  Fangzhen Lin,et al.  Situation Calculus , 2008, Handbook of Knowledge Representation.

[12]  S. Coles,et al.  Modelling Association Football Scores and Inefficiencies in the Football Betting Market , 1997 .

[13]  Jesse Davis,et al.  VAEP: An Objective Approach to Valuing On-the-Ball Actions in Soccer (Extended Abstract) , 2020, IJCAI.

[14]  Bo An,et al.  PROTECT: An Application of Computational Game Theory for the Security of the Ports of the United States , 2012, AAAI.

[15]  Sarvapali D. Ramchurn,et al.  Artificial intelligence for team sports: a survey , 2019, The Knowledge Engineering Review.

[16]  Tom Rodden,et al.  A Disaster Response System based on Human-Agent Collectives , 2015, J. Artif. Intell. Res..

[17]  M. Paczuski,et al.  Price Variations in a Stock Market with Many Agents , 1997 .

[18]  Sarvapali D. Ramchurn,et al.  Optimising Game Tactics for Football , 2020, AAMAS.

[19]  Sarit Kraus,et al.  Playing games for security: an efficient exact algorithm for solving Bayesian Stackelberg games , 2008, AAMAS.

[20]  Sarvapali D. Ramchurn,et al.  Learning the Value of Teamwork to Form Efficient Teams , 2020, AAAI.

[21]  Olivier Pietquin,et al.  Observational Learning by Reinforcement Learning , 2017, AAMAS.

[22]  Sungzoon Cho,et al.  Ensemble learning using observational learning theory , 1999, IJCNN'99. International Joint Conference on Neural Networks. Proceedings (Cat. No.99CH36339).

[23]  Demis Hassabis,et al.  Mastering the game of Go with deep neural networks and tree search , 2016, Nature.