Paperlane

no. 16 · Tuesday 17 November 2026

Barron Approximation and locally optimal weight densities for shallow neural networks

G. Welper

math.NA · via cs.LG

Abstract

Mean field theories provide optimization results for shallow neural networks by analyzing the weight distribution in infinite width limits. Corresponding results for finite sized networks are obtained by particle approximations, for which sharp quantitative bounds are still an open problem. In this paper, we consider a modified mean field loss, which allows a more fine grained control over finite sized networks. We prove convexity and equidistribution properties, which directly lead to Barron type approximation results for networks sampled from local loss minimizers. We demonstrate that particle approximations of the new loss function naturally lead to gradient descent methods with dropout type regularization.

fields
Numerical Analysis
pdf ↗ page arXiv:2611.19798 ↗