The Application of Reinforcement Learning to the Simple Soccer Game

Abstract

This thesis looks at Reinforcement Learning (RL) and its application to computer-simulated soccer games. RL itself is explained and its main points detailed. A summary of research into computer simulated soccer is then outlined. The Simple Soccer game is introduced and an RL technique (tabular Q-learning) is applied to it. Aspects of the research literature are reviewed in the context of Simple Soccer, such as the issues involved in designing and constructing RL experiments using games. Finally the results of the application are presented and an analysis of the findings is made