Abstract

ABSTRACT A novel model of associative memory with biorthogonal properties is presented in this paper, which can be viewedas an improved version of T. Kohonen's linear model of associative memory. An iterative algorithm is developed, whichmakes our model directly employed without any limit condition. Several characteristics of our model are also discussed, which are very similar to biological phenomena. It is shown that the optimal value of an associative memory can be always obtained in our model. Some conclusions remarks with respect to this model are also given.keywords: Neural networks, associative memory, biorthogonal properties, iterative algorithm. 1. INTRODUCTIONThe fundamental mechanisms and functions of human memory are an important issue of psychology and physiol-ogy etc., and there is a long studying history about them[1]. Recently, these problems are more and more noted by Alresearchers to pursue their goal. Moreover, the study of memory has become a component of Al research. There are

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call