Learning Strategies from Reinforcement Learning Agents
How does AI play football? An analysis of RL and real-world football strategies
Reinforcement learning (rl) has made it possible to develop sophisticated agents that excel in a wide range of applications.
Simulations using such agents can provide valuable information in scenariosthat are difficult to scientifically experiment in the real world.
In this paper, we explore what can be learnt from the use of simulated environments by using aggregated statisticsand social network analysis (sna).
We found that (1) there are strong correlations between the competitiveness of an agent and various snametrics and (2) aspects of the rl agents play style become similar to realworld footballers as the agent becomes more competitive.