Off-Beat Multi-Agent Reinforcement Learning

We investigate cooperative multi-agent reinforcement learning in environments with off-beat actions, i. e. , all actions have execution durations. During execution durations, the environmental changes are not synchronised with action executions. To learn efficient multi-agent coordination in environments with off-beat actions, we propose a novel reward redistribution method built on our novel graph-based episodic memory. We name our solution method as LeGEM. Empirical results on stag-hunter game show that it significantly boosts multi-agent coordination.

Authors

• Wei Qiu
• Weixun Wang Netease Fuxi AI Lab
• Rundong Wang
• Bo An Nanyang Technological University
• Yujing Hu NetEase Fuxi AI Lab, Hangzhou, China
• Svetlana Obraztsova Carleton University
• Zinovi Rabinovich Carleton University
• Jianye Hao Tianjin University
• Yingfeng Chen
• Changjie Fan Fuxi AI Lab, NetEase, Inc.

Keywords

• multi-agent coordination
• multi-agent reinforcement learning