[1E5-GS-5] Agents:

[1E5-GS-5-04] Adaptive action utilization in reinforcement learning from real-world multi-agent demonstrations

〇Keisuke Fujii^1,2,5, Kazushi Tsutusi¹, Atom Scott¹, Hiroshi Nakahara¹, Naoya Takeishi^3,2, Yoshinobu Kawahara^4,2 (1. Nagoya University, 2. RIKEN, 3. The University of Tokyo, 4. Osaka University, 5. JST Presto)

Keywords:Reinforcement learning, Machine Learning, Sports, Deep Learning

When modeling real-world biological multi-agents with reinforcement learning, there is a domain gap between the source real-world data and the target reinforcement learning environment. Therefore, the target dynamics are adapted to the unknown source dynamics. In this study, we propose a reinforcement learning method that uses information obtained by adapting source action to target action in a supervised manner as a method for domain adaptation in multi-agent reinforcement learning from real-world demonstrations. In limited situations such as 2vs1 chase-escape, 2vs2 and 4vs8 in soccer, we show that the agent learned to imitate the demonstrations and obtain rewards compared to the baseline.