Reinforcement Learning Based on Active Learning Method

TL;DR

This paper introduces a reinforcement learning method that integrates Active Learning Method (ALM) with an actor-critic framework, utilizing fuzzy modeling and TD learning to improve control system behavior adaptation.

Contribution

It presents a novel reinforcement learning approach combining ALM with actor-critic architecture, capable of learning with or without predefined fuzzy systems.

Findings

System can learn effectively using ALM-based reinforcement learning.

The approach adapts to control tasks with delayed reinforcement signals.

Learning performance is demonstrated through simulation results.

Abstract

In this paper, a new reinforcement learning approach is proposed which is based on a powerful concept named Active Learning Method (ALM) in modeling. ALM expresses any multi-input-single-output system as a fuzzy combination of some single-input-singleoutput systems. The proposed method is an actor-critic system similar to Generalized Approximate Reasoning based Intelligent Control (GARIC) structure to adapt the ALM by delayed reinforcement signals. Our system uses Temporal Difference (TD) learning to model the behavior of useful actions of a control system. The goodness of an action is modeled on Reward- Penalty-Plane. IDS planes will be updated according to this plane. It is shown that the system can learn with a predefined fuzzy system or without it (through random actions).