Finite-dimensional Gaussian approximation for deep neural networks: universality in random weights

DOI10.3150/26-BEJ1978MaRDI QIDQ7347143zbMATH OpenFDO

Authors Krishnakumar Balasubramanian, Nathan Ross

Publication date 14 September 2026

Published in Bernoulli (Search for Journal in Brave)

Full work available at URL https://projecteuclid.org/journals/bernoulli/volume-32/issue-4/Finite-dimensional-Gaussian-approximation-for-deep-neural-networks-Universality/10.3150/26-BEJ1978.full

Mathematics Subject Classification ID

Central limit and other weak theorems (60F05) Gaussian processes (60G15) Artificial neural networks and deep learning (68T07)

Cites work

• -Stable convergence of heavy-/light-tailed infinitely wide neural networks
• A Quantitative Functional Central Limit Theorem for Shallow Neural Networks
• Bayesian learning for neural networks
• Deep stable neural networks: large-width asymptotics and convergence rates
• Gaussian random field approximation via Stein's method with applications to wide random neural networks
• Multivariate normal approximation using exchangeable pairs
• Multivariate normal approximation with Stein's method of exchangeable pairs under a general linearity condition
• Multivariate normal approximations by Stein's method and size bias couplings
• Normal Approximation by Stein’s Method
• Normal approximation of random Gaussian neural networks
• Normal approximations with Malliavin calculus. From Stein's method to universality
• On Stein's method for multivariate normal approximation
• On the rate of convergence in the multivariate CLT
• Quantitative CLTs in deep neural networks
• Random neural networks in the infinite width limit as Gaussian processes

This page was built for publication: Finite-dimensional Gaussian approximation for deep neural networks: universality in random weights

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q7347143)