Extension of Learning by Switching Rule-Generation and Reasoning Methods - Acquisition of Meta-rules for Switching with Reinforcement Learning

Abstract

When we solve a problem, we firstly have no knowledge and gradually acquire some piece of knowledge by observing new data,and at last arrive at complete knowledge for solving the problem.To implement such kind of learning mechanism, we proposed a learning method of switching reasoning methods and rule generation methods. In this method, meta-rules for switching was given apriori. In this paper, we propose a method that aquire them from the number of reasonings and the successive number of incorrect reasonings with reinforcement learning.