Afshin Oroojlooy, SAS
Pegah Rahimian, Budapest University of Technology and Economics
Laszlo Toka, MTA-BME Information Systems Research Group
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.
At SAS, we take big ideas that shape the future and make them bigger. On these pages we've collected the best examples of analytical and technology research at SAS, with a spotlight on the talented people that make it possible.
SAS doesn't settle for ordinary innovation. We put thought and research into ideas that matter – and then create analytics solutions that improve business and society.
Work with us. Visit sas.com/careers to find R&D opportunities at SAS.
Read more stories at the Data Science Resource Hub
More science and SAS: SAS authors who contribute to Health & Life Science research, and use of SAS software in scientific research.