Saved in:
Bibliographic Details
Main Authors: Vivet, Arnau, Arenas, Alex
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2604.25481
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910173358981120
author Vivet, Arnau
Arenas, Alex
author_facet Vivet, Arnau
Arenas, Alex
contents We introduce a Hopfield-type associative memory in which effective connectivity is multiplicatively modulated by astrocytic gains evolving under an entropy-regularized replicator equation. The coupled neuron-astrocyte dynamics admit a Lyapunov function, ensuring global convergence. At fixed points, astrocytic gains implement a softmax-normalized allocation over pattern similarity scores, yielding a mechanistic realization of self-attention as emergent routing on the gain simplex. In regimes of high memory load and interference, the model significantly improves retrieval accuracy relative to classical Hopfield dynamics and recent neuron-astrocyte baselines. These results establish a dynamical systems framework linking glial modulation, competitive resource allocation, and attention-like computation.
format Preprint
id arxiv_https___arxiv_org_abs_2604_25481
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Emergent Self-Attention from Astrocyte-Gated Associative Memory Dynamics
Vivet, Arnau
Arenas, Alex
Data Analysis, Statistics and Probability
Machine Learning
Adaptation and Self-Organizing Systems
Physics and Society
We introduce a Hopfield-type associative memory in which effective connectivity is multiplicatively modulated by astrocytic gains evolving under an entropy-regularized replicator equation. The coupled neuron-astrocyte dynamics admit a Lyapunov function, ensuring global convergence. At fixed points, astrocytic gains implement a softmax-normalized allocation over pattern similarity scores, yielding a mechanistic realization of self-attention as emergent routing on the gain simplex. In regimes of high memory load and interference, the model significantly improves retrieval accuracy relative to classical Hopfield dynamics and recent neuron-astrocyte baselines. These results establish a dynamical systems framework linking glial modulation, competitive resource allocation, and attention-like computation.
title Emergent Self-Attention from Astrocyte-Gated Associative Memory Dynamics
topic Data Analysis, Statistics and Probability
Machine Learning
Adaptation and Self-Organizing Systems
Physics and Society
url https://arxiv.org/abs/2604.25481