The fixed point assignment problem in neural networks and its application to associative memory

A problem of assigning a prescribed set of vectors to asymptotically stable fixed points of a system arises from constructing associative memory using a neural network. This paper deals with this problem and discusses a method for constructing a neural network which satisfies the properties that not only a prescribed set of vectors is assigned to its fixed points but also each fixed point achieves a maximum convergence margin to improve the capability as associative memory. Finally to illustrate the result a simple numerical example is worked out.