Behavior acquisition by multi-layered reinforcement learning

Abstract

This paper proposes multi-layered reinforcement learning by which the control structure can be decomposed into smaller transportable chunks and therefore previously learned knowledge can be applied to related tasks in a newly encountered situations. The modules in the lower networks are organized as experts to move into different categories of sensor output regions and to learn lower level behaviors using motor commands. In the meantime, the modules in the higher networks are organized as experts which learn higher level behavior using lower modules. We apply the method to a simple soccer situation in the context of RoboCup, show the experimental results, and give a discussion.