LDP for the covariance process in fully connected neural networks

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Andreis, Luisa, Bassetti, Federico, Hirsch, Christian
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909608551907328
author Andreis, Luisa
Bassetti, Federico
Hirsch, Christian
author_facet Andreis, Luisa
Bassetti, Federico
Hirsch, Christian
contents In this work, we study large deviation properties of the covariance process in fully connected Gaussian deep neural networks. More precisely, we establish a large deviation principle (LDP) for the covariance process in a functional framework, viewing it as a process in the space of continuous functions. As key applications of our main results, we obtain posterior LDPs under Gaussian likelihood in both the infinite-width and mean-field regimes. The proof is based on an LDP for the covariance process as a Markov process valued in the space of non-negative, symmetric trace-class operators equipped with the trace norm.
format Preprint
id arxiv_https___arxiv_org_abs_2505_08062
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle LDP for the covariance process in fully connected neural networks
Andreis, Luisa
Bassetti, Federico
Hirsch, Christian
Probability
60F10, 60G15, 62E2, 68T07
In this work, we study large deviation properties of the covariance process in fully connected Gaussian deep neural networks. More precisely, we establish a large deviation principle (LDP) for the covariance process in a functional framework, viewing it as a process in the space of continuous functions. As key applications of our main results, we obtain posterior LDPs under Gaussian likelihood in both the infinite-width and mean-field regimes. The proof is based on an LDP for the covariance process as a Markov process valued in the space of non-negative, symmetric trace-class operators equipped with the trace norm.
title LDP for the covariance process in fully connected neural networks
topic Probability
60F10, 60G15, 62E2, 68T07
url https://arxiv.org/abs/2505.08062