Multi-agent Case-Based Reasoning for Cooperative Reinforcement Learners

摘要

In both research fields, Case-Based Reasoning and Reinforcement Learning, the system under consideration gains its expertise from experience. Utilizing this fundamental common ground as well as further characteristics and results of these two disciplines, in this paper we develop an approach that facilitates the distributed learning of behaviour policies in cooperative multi-agent domains without communication between the learning agents. We evaluate our algorithms in a case study in reactive production scheduling.