Vector-Neuron Models of Associative Memory

TL;DR

This paper introduces two advanced models of associative memory with multi-state neurons, demonstrating superior storage capacity and noise immunity compared to traditional Hopfield networks, and provides a unified formalism for their description.

Contribution

The paper presents novel multi-state neuron models, PGNN and PNN, with enhanced storage and noise resistance, along with a unified formalism for their analysis.

Findings

Models surpass Hopfield network in capacity and noise immunity

Unified formalism for PNN and PGNN developed

Inherent mechanisms for recognition properties clarified

Abstract

We consider two models of Hopfield-like associative memory with $q$-valued neurons: Potts-glass neural network (PGNN) and parametrical neural network (PNN). In these models neurons can be in more than two different states. The models have the record characteristics of its storage capacity and noise immunity, and significantly exceed the Hopfield model. We present a uniform formalism allowing us to describe both PNN and PGNN. This networks inherent mechanisms, responsible for outstanding recognizing properties, are clarified.