Thermodynamics of bidirectional associative memories

TL;DR

This paper analyzes the thermodynamic properties of bidirectional associative memories (BAMs), revealing their phase diagram, storage capacity, and retrieval mechanisms, and compares their efficiency to Hopfield models using statistical physics techniques.

Contribution

It provides a rigorous thermodynamic analysis of BAMs, including phase diagrams and capacity, and explores their relation to Hopfield models and Restricted Boltzmann Machines.

Findings

BAMs can store more efficiently with layer asymmetry.

Critical load depends on noise, load, and asymmetry.

Retrieval mechanism resembles interacting Hopfield models.

Abstract

In this paper we investigate the equilibrium properties of bidirectional associative memories (BAMs). Introduced by Kosko in 1988 as a generalization of the Hopfield model to a bipartite structure, the simplest architecture is defined by two layers of neurons, with synaptic connections only between units of different layers: even without internal connections within each layer, information storage and retrieval are still possible through the reverberation of neural activities passing from one layer to another. We characterize the computational capabilities of a stochastic extension of this model in the thermodynamic limit, by applying rigorous techniques from statistical physics. A detailed picture of the phase diagram at the replica symmetric level is provided, both at finite temperature and in the noiseless regimes. Also for the latter, the critical load is further investigated up to one step of replica symmetry breaking. An analytical and numerical inspection of the transition curves (namely critical lines splitting the various modes of operation of the machine) is carried out as the control parameters - noise, load and asymmetry between the two layer sizes - are tuned. In particular, with a finite asymmetry between the two layers, it is shown how the BAM can store information more efficiently than the Hopfield model by requiring less parameters to encode a fixed number of patterns. Comparisons are made with numerical simulations of neural dynamics. Finally, a low-load analysis is carried out to explain the retrieval mechanism in the BAM by analogy with two interacting Hopfield models. A potential equivalence with two coupled Restricted Boltmzann Machines is also discussed.