Dynamical stability for dense patterns in discrete attractor neural networks

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Cohen, Uri, Lengyel, Máté
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866918299195932672
author Cohen, Uri
Lengyel, Máté
author_facet Cohen, Uri
Lengyel, Máté
contents Neural networks storing multiple discrete attractors are canonical models of biological memory. Previously, the dynamical stability of such networks could only be guaranteed under highly restrictive conditions. Here, we derive a theory of the local stability of discrete fixed points in a broad class of networks with graded neural activities and in the presence of noise. By directly analyzing the bulk and the outliers of the Jacobian spectrum, we show that all fixed points are stable below a critical load that is distinct from the classical \textit{critical capacity} and depends on the statistics of neural activities in the fixed points as well as the single-neuron activation function. Our analysis highlights the computational benefits of threshold-linear activation and sparse-like patterns.
format Preprint
id arxiv_https___arxiv_org_abs_2507_10383
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Dynamical stability for dense patterns in discrete attractor neural networks
Cohen, Uri
Lengyel, Máté
Disordered Systems and Neural Networks
Statistical Mechanics
Machine Learning
Neural and Evolutionary Computing
Neurons and Cognition
Neural networks storing multiple discrete attractors are canonical models of biological memory. Previously, the dynamical stability of such networks could only be guaranteed under highly restrictive conditions. Here, we derive a theory of the local stability of discrete fixed points in a broad class of networks with graded neural activities and in the presence of noise. By directly analyzing the bulk and the outliers of the Jacobian spectrum, we show that all fixed points are stable below a critical load that is distinct from the classical \textit{critical capacity} and depends on the statistics of neural activities in the fixed points as well as the single-neuron activation function. Our analysis highlights the computational benefits of threshold-linear activation and sparse-like patterns.
title Dynamical stability for dense patterns in discrete attractor neural networks
topic Disordered Systems and Neural Networks
Statistical Mechanics
Machine Learning
Neural and Evolutionary Computing
Neurons and Cognition
url https://arxiv.org/abs/2507.10383