Saved in:
Bibliographic Details
Main Authors: Aguilera, Miguel, Morales, Pablo A., Rosas, Fernando E., Shimazaki, Hideaki
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2408.02326
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908700667543552
author Aguilera, Miguel
Morales, Pablo A.
Rosas, Fernando E.
Shimazaki, Hideaki
author_facet Aguilera, Miguel
Morales, Pablo A.
Rosas, Fernando E.
Shimazaki, Hideaki
contents Higher-order interactions underlie complex phenomena in systems such as biological and artificial neural networks, but their study is challenging due to the scarcity of tractable models. By leveraging a generalisation of the maximum entropy principle, we introduce curved neural networks as a class of models with a limited number of parameters that are particularly well-suited for studying higher-order phenomena. Through exact mean-field descriptions, we show that these curved neural networks implement a self-regulating annealing process that can accelerate memory retrieval, leading to explosive order-disorder phase transitions with multi-stability and hysteresis effects. Moreover, by analytically exploring their memory-retrieval capacity using the replica trick, we demonstrate that these networks can enhance memory capacity and robustness of retrieval over classical associative-memory networks. Overall, the proposed framework provides parsimonious models amenable to analytical study, revealing higher-order phenomena in complex networks.
format Preprint
id arxiv_https___arxiv_org_abs_2408_02326
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Explosive neural networks via higher-order interactions in curved statistical manifolds
Aguilera, Miguel
Morales, Pablo A.
Rosas, Fernando E.
Shimazaki, Hideaki
Disordered Systems and Neural Networks
Statistical Mechanics
Information Theory
Adaptation and Self-Organizing Systems
Machine Learning
Higher-order interactions underlie complex phenomena in systems such as biological and artificial neural networks, but their study is challenging due to the scarcity of tractable models. By leveraging a generalisation of the maximum entropy principle, we introduce curved neural networks as a class of models with a limited number of parameters that are particularly well-suited for studying higher-order phenomena. Through exact mean-field descriptions, we show that these curved neural networks implement a self-regulating annealing process that can accelerate memory retrieval, leading to explosive order-disorder phase transitions with multi-stability and hysteresis effects. Moreover, by analytically exploring their memory-retrieval capacity using the replica trick, we demonstrate that these networks can enhance memory capacity and robustness of retrieval over classical associative-memory networks. Overall, the proposed framework provides parsimonious models amenable to analytical study, revealing higher-order phenomena in complex networks.
title Explosive neural networks via higher-order interactions in curved statistical manifolds
topic Disordered Systems and Neural Networks
Statistical Mechanics
Information Theory
Adaptation and Self-Organizing Systems
Machine Learning
url https://arxiv.org/abs/2408.02326