Mutual information and self-control of a fully-connected low-activity neural network

TL;DR

This paper investigates a self-control mechanism in a fully-connected three-state neural network, showing that a dynamic threshold improves mutual information, storage capacity, and retrieval robustness.

Contribution

It introduces a time-dependent threshold based on neural and pattern activity, enhancing network performance in low-activity regimes.

Findings

Self-control improves mutual information content.

Storage capacity increases with self-adaptation.

Basins of attraction are enlarged due to the self-control mechanism.

Abstract

A self-control mechanism for the dynamics of a three-state fully-connected neural network is studied through the introduction of a time-dependent threshold. The self-adapting threshold is a function of both the neural and the pattern activity in the network. The time evolution of the order parameters is obtained on the basis of a recently developed dynamical recursive scheme. In the limit of low activity the mutual information is shown to be the relevant parameter in order to determine the retrieval quality. Due to self-control an improvement of this mutual information content as well as an increase of the storage capacity and an enlargement of the basins of attraction are found. These results are compared with numerical simulations.