The BSB neural network in the convex body spanned by the prototype patterns for associative memory

Abstract

This paper presents a kind of general brain-state-in-a-box (BSB) neural network for associative memory. It can guarantee that the set of prototype patterns is the same as the set of asymptotically stable equilibrium points, an equilibrium point which is not asymptotically stable is just the state that cannot be recognized. Furthermore, the basin of attraction of each desired memory pattern is distributed reasonably (in Hamming distance sense). The proposed network improves classical BSB as an idea model for associative memory. The network considered herein is a generalization of traditional BSB, its state is allowed to lie in a general closed convex set, which is spanned by the prototype patterns. The performance of the proposed network is demonstrated by means of simulation of one numerical example.