Modular fuzzy-reinforcement learning in multi-agent systems

Abstract

In multi-agent systems, the state space to be handled constitutes a major problem efficiently in learning of agents, This paper presents a novel approach to overcome this problem. The approach uses together the advantages of the modular architecture and fuzzy logic in multi-agent systems. Fuzzy logic maps the input fuzzy sets, representing state space of each learning module, to the output fuzzy sets representing the action space. The fuzzy rule base of each module is built through the Q-learning, which is one of the reinforcement learning schemes, Experimental results done on pursuit domain show the effectiveness and applicability of the proposed approach.