Behavior Patterns of adaptive Multi-Joined Robot learned by Multi-Agent Influence Reinforcement Learning

Abstract

This paper describes behavior patterns produced by Multi-Joined Robot learned via Influence Reinforcement learning. This learning technique used for distributed, adaptive and self-organizing control in multi-agent system. This technique is quite simple and uses agent’s influences to estimate learning error between them. As will show, this learning rule supports positive-reward interactions between agents and does not require any additional information than standard reinforcement learning. The behavior patterns of learned robot shows that optimal behavior strategies differ for various learning techniques. As we will show, every algorithm produces his own behavior's patterns which are optimal for that learning rule to produce a faster convergence.