Adaptive action supervision in reinforcement learning from real-world multi-agent demonstrations
This work addresses the challenge of applying RL to real-world multi-agent systems with unknown dynamics, offering a method to improve behavior modeling in scientific and engineering fields, though it appears incremental as it builds on existing RL and supervised learning techniques.
The paper tackles the problem of modeling real-world biological multi-agents in reinforcement learning (RL) by addressing the domain gap between real-world demonstrations and cyberspace environments, proposing an adaptive action supervision method that balances reproducibility and generalization, achieving successful performance in tasks like football with professional player data.
Modeling of real-world biological multi-agents is a fundamental problem in various scientific and engineering fields. Reinforcement learning (RL) is a powerful framework to generate flexible and diverse behaviors in cyberspace; however, when modeling real-world biological multi-agents, there is a domain gap between behaviors in the source (i.e., real-world data) and the target (i.e., cyberspace for RL), and the source environment parameters are usually unknown. In this paper, we propose a method for adaptive action supervision in RL from real-world demonstrations in multi-agent scenarios. We adopt an approach that combines RL and supervised learning by selecting actions of demonstrations in RL based on the minimum distance of dynamic time warping for utilizing the information of the unknown source dynamics. This approach can be easily applied to many existing neural network architectures and provide us with an RL model balanced between reproducibility as imitation and generalization ability to obtain rewards in cyberspace. In the experiments, using chase-and-escape and football tasks with the different dynamics between the unknown source and target environments, we show that our approach achieved a balance between the reproducibility and the generalization ability compared with the baselines. In particular, we used the tracking data of professional football players as expert demonstrations in football and show successful performances despite the larger gap between behaviors in the source and target environments than the chase-and-escape task.