Multitasking associative networks

TL;DR

This paper presents a new bipartite, diluted neural network model that can retrieve multiple patterns simultaneously without errors, offering insights into multitasking capabilities of biological systems.

Contribution

It introduces a novel sparse restricted Boltzmann machine model with thermodynamic equivalence to associative memory, enabling parallel pattern retrieval without spurious states.

Findings

Model successfully retrieves multiple patterns in parallel.

Results align across statistical mechanics, signal-to-noise, and simulations.

Provides biological insights into multitasking in neural and immune networks.

Abstract

We introduce a bipartite, diluted and frustrated, network as a sparse restricted Boltzman machine and we show its thermodynamical equivalence to an associative working memory able to retrieve multiple patterns in parallel without falling into spurious states typical of classical neural networks. We focus on systems processing in parallel a finite (up to logarithmic growth in the volume) amount of patterns, mirroring the low-level storage of standard Amit-Gutfreund-Sompolinsky theory. Results obtained trough statistical mechanics, signal-to-noise technique and Monte Carlo simulations are overall in perfect agreement and carry interesting biological insights. Indeed, these associative networks pave new perspectives in the understanding of multitasking features expressed by complex systems, e.g. neural and immune networks.