How Soar Agent Learns: Episodic Memory

Abstract

The main objective of this paper is to make the agent learn from its experiences in order to improve its intelligence. Learning is performed with the help of episodic memory which store the previous experiences in the form of episodes in order to solve the problem. With the help of the implementation in soar of an example, this paper presents the improvement in intelligence. In this paper episodes (previous experiences) which are in the form of WME (working memory elements) work to improve the intelligence. Objective of this paper is to show that learning always improves performance i.e. as agent learns more, the performance will improve.