Authors:
Keisuke Fujii
1
;
2
;
3
;
Kazushi Tsutsui
1
;
Atom Scott
1
;
Hiroshi Nakahara
1
;
Naoya Takeishi
2
;
4
and
Yoshinobu Kawahara
2
;
5
Affiliations:
1
Graduate School of Informatics, Nagoya University, Nagoya, Aichi, Japan
;
2
Center for Advanced Intelligence Project, RIKEN, Osaka, Osaka, Japan
;
3
PRESTO, Japan Science and Technology Agency, Tokyo, Japan
;
4
The Graduate School of Engineering, The University of Tokyo, Tokyo, Japan
;
5
Graduate School of Information Science and Technology, Osaka University, Osaka, Japan
Keyword(s):
Neural Networks, Trajectory, Simulation, Multi-Agent.
Abstract:
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.
(More)