A spectral approach to Hebbian-like neural networks

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Agliari, Elena, Luongo, Domenico, Fachechi, Alberto
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909610651156480
author Agliari, Elena
Luongo, Domenico
Fachechi, Alberto
author_facet Agliari, Elena
Luongo, Domenico
Fachechi, Alberto
contents We consider the Hopfield neural network as a model of associative memory and we define its neuronal interaction matrix $\mathbf{J}$ as a function of a set of $K \times M$ binary vectors $\{\mathbfξ^{μ, A} \}_{μ=1,...,K}^{A=1,...,M}$ representing a sample of the reality that we want to retrieve. In particular, any item $\mathbfξ^{μ, A}$ is meant as a corrupted version of an unknown ground pattern $\mathbfζ^μ$, that is the target of our retrieval process. We consider and compare two definitions for $\mathbf{J}$, referred to as supervised and unsupervised, according to whether the class $μ$, each example belongs to, is unveiled or not, also, these definitions recover the paradigmatic Hebb's rule under suitable limits. The spectral properties of the resulting matrices are studied and used to inspect the retrieval capabilities of the related models as a function of their control parameters.
format Preprint
id arxiv_https___arxiv_org_abs_2401_16114
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle A spectral approach to Hebbian-like neural networks
Agliari, Elena
Luongo, Domenico
Fachechi, Alberto
Mathematical Physics
Disordered Systems and Neural Networks
We consider the Hopfield neural network as a model of associative memory and we define its neuronal interaction matrix $\mathbf{J}$ as a function of a set of $K \times M$ binary vectors $\{\mathbfξ^{μ, A} \}_{μ=1,...,K}^{A=1,...,M}$ representing a sample of the reality that we want to retrieve. In particular, any item $\mathbfξ^{μ, A}$ is meant as a corrupted version of an unknown ground pattern $\mathbfζ^μ$, that is the target of our retrieval process. We consider and compare two definitions for $\mathbf{J}$, referred to as supervised and unsupervised, according to whether the class $μ$, each example belongs to, is unveiled or not, also, these definitions recover the paradigmatic Hebb's rule under suitable limits. The spectral properties of the resulting matrices are studied and used to inspect the retrieval capabilities of the related models as a function of their control parameters.
title A spectral approach to Hebbian-like neural networks
topic Mathematical Physics
Disordered Systems and Neural Networks
url https://arxiv.org/abs/2401.16114