Approximation Rates for Shallow ReLU$^k$ Neural Networks on Sobolev Spaces via the Radon Transform

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Mao, Tong, Siegel, Jonathan W., Xu, Jinchao
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908597759246336
author Mao, Tong
Siegel, Jonathan W.
Xu, Jinchao
author_facet Mao, Tong
Siegel, Jonathan W.
Xu, Jinchao
contents Let $Ω\subset \mathbb{R}^d$ be a bounded domain. We consider the problem of how efficiently shallow neural networks with the ReLU$^k$ activation function can approximate functions from Sobolev spaces $W^s(L_p(Ω))$ with error measured in the $L_q(Ω)$-norm. Utilizing the Radon transform and recent results from discrepancy theory, we provide a simple proof of nearly optimal approximation rates in a variety of cases, including when $q\leq p$, $p\geq 2$, and $s \leq k + (d+1)/2$. The rates we derive are optimal up to logarithmic factors, and significantly generalize existing results. An interesting consequence is that the adaptivity of shallow ReLU$^k$ neural networks enables them to obtain optimal approximation rates for smoothness up to order $s = k + (d+1)/2$, even though they represent piecewise polynomials of fixed degree $k$.
format Preprint
id arxiv_https___arxiv_org_abs_2408_10996
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Approximation Rates for Shallow ReLU$^k$ Neural Networks on Sobolev Spaces via the Radon Transform
Mao, Tong
Siegel, Jonathan W.
Xu, Jinchao
Machine Learning
Numerical Analysis
62M45, 41A25, 41A30
Let $Ω\subset \mathbb{R}^d$ be a bounded domain. We consider the problem of how efficiently shallow neural networks with the ReLU$^k$ activation function can approximate functions from Sobolev spaces $W^s(L_p(Ω))$ with error measured in the $L_q(Ω)$-norm. Utilizing the Radon transform and recent results from discrepancy theory, we provide a simple proof of nearly optimal approximation rates in a variety of cases, including when $q\leq p$, $p\geq 2$, and $s \leq k + (d+1)/2$. The rates we derive are optimal up to logarithmic factors, and significantly generalize existing results. An interesting consequence is that the adaptivity of shallow ReLU$^k$ neural networks enables them to obtain optimal approximation rates for smoothness up to order $s = k + (d+1)/2$, even though they represent piecewise polynomials of fixed degree $k$.
title Approximation Rates for Shallow ReLU$^k$ Neural Networks on Sobolev Spaces via the Radon Transform
topic Machine Learning
Numerical Analysis
62M45, 41A25, 41A30
url https://arxiv.org/abs/2408.10996