Learning to Teach in Cooperative Multiagent Reinforcement Learning

• DOI: 10.1609/aaai.v33i01.33016128
• Publication Date: 2019-08-27T07:48:34Z
• Authors: Shayegan Omidshafiei; Dong-Ki Kim; Miao Liu; Gerald Tesauro; Matthew Riemer; Christopher Amato; Murray Campbell; Jonathan P. How
• Keywords: FOS: Computer and information sciences; Artificial Intelligence (cs.AI); Computer Science - Artificial Intelligence; 0202 electrical engineering, electronic engineering, information engineering; Computer Science - Multiagent Systems; 02 engineering and technology; Multiagent Systems (cs.MA)

Collective human knowledge has clearly benefited from the fact that innovations by individuals are taught to others through communication. Similar social groups, agents in distributed learning systems would likely benefit communication share and teach skills. The problem of teaching improve agent been investigated prior works, but these approaches make assumptions prevent application general multiagent problems, or require domain expertise for problems they can apply to. This inherent complexities related measuring long-term impacts compound standard coordination challenges. In contrast existing this paper presents first framework algorithm intelligent learn a environment. Our algorithm, Learning Coordinate Teach Reinforcement (LeCTR), addresses peer-to-peer cooperative reinforcement learning. Each our approach learns both when what advise, then uses received advice local Importantly, roles not fixed; assume role student and/or teacher at appropriate moments, requesting providing order teamwide performance Empirical comparisons against state-of-the-art methods show only significantly faster, also coordinate tasks where fail.