Concentration of measure for non-linear random matrices with applications to neural networks and non-commutative polynomials

Fuente: arXiv
Saved in:
Bibliographic Details
Main Author: Adamczak, Radosław
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908448621330432
author Adamczak, Radosław
author_facet Adamczak, Radosław
contents We prove concentration inequalities for several models of non-linear random matrices. As corollaries we obtain estimates for linear spectral statistics of the conjugate kernel of neural networks and non-commutative polynomials in (possibly dependent) random matrices.
format Preprint
id arxiv_https___arxiv_org_abs_2507_07625
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Concentration of measure for non-linear random matrices with applications to neural networks and non-commutative polynomials
Adamczak, Radosław
Probability
Machine Learning
Primary: 60B20, 60E15, Secondary: 68T07
We prove concentration inequalities for several models of non-linear random matrices. As corollaries we obtain estimates for linear spectral statistics of the conjugate kernel of neural networks and non-commutative polynomials in (possibly dependent) random matrices.
title Concentration of measure for non-linear random matrices with applications to neural networks and non-commutative polynomials
topic Probability
Machine Learning
Primary: 60B20, 60E15, Secondary: 68T07
url https://arxiv.org/abs/2507.07625