A novel Episodic Associative Memory model for enhanced classification accuracy

Abstract

A novel approach to Episodic Associative Memory (EAM), known as Episodic Associative Memory with a Neighborhood Effect (EAMwNE) is presented in this paper. It overcomes the representation limitations of existing episodic memory models and increases the potential for their use in practical application. © 2007 Elsevier B.V. All rights reserved.