Parallel retrieval of correlated patterns

TL;DR

This paper introduces a novel associative network model that combines correlated attractor and multitasking extensions of the Hopfield model, enabling parallel pattern retrieval with tunable quality based on pattern correlations.

Contribution

The authors develop a merged model of correlated and multitasking Hopfield networks, analyzed through statistical mechanics and Hamilton-Jacobi interpolation, demonstrating thermodynamic equivalence with a Boltzmann machine.

Findings

The model can retrieve multiple patterns in parallel with quality depending on pattern correlation.

Different parameter settings produce hierarchical or symmetric retrieval outputs.

The approach is validated by numerical simulations and analytical methods.

Abstract

In this work, we first revise some extensions of the standard Hopfield model in the low storage limit, namely the correlated attractor case and the multitasking case recently introduced by the authors. The former case is based on a modification of the Hebbian prescription, which induces a coupling between consecutive patterns and this effect is tuned by a parameter $a$. In the latter case, dilution is introduced in pattern entries, in such a way that a fraction $d$ of them is blank. Then, we merge these two extensions to obtain a system able to retrieve several patterns in parallel and the quality of retrieval, encoded by the set of Mattis magnetizations ${m^{\mu}}$, is reminiscent of the correlation among patterns. By tuning the parameters $d$ and $a$, qualitatively different outputs emerge, ranging from highly hierarchical, to symmetric. The investigations are accomplished by means of both numerical simulations and statistical mechanics analysis, properly adapting a novel technique originally developed for spin glasses, i.e. the Hamilton-Jacobi interpolation, with excellent agreement. Finally, we show the thermodynamical equivalence of this associative network with a (restricted) Boltzmann machine and study its stochastic dynamics to obtain even a dynamical picture, perfectly consistent with the static scenario earlier discussed.