ReLU neural network approximation to piecewise constant functions

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Cai, Zhiqiang, Choi, Junpyo, Liu, Min
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866929553982619648
author Cai, Zhiqiang
Choi, Junpyo
Liu, Min
author_facet Cai, Zhiqiang
Choi, Junpyo
Liu, Min
contents This paper studies the approximation property of ReLU neural networks (NNs) to piecewise constant functions with unknown interfaces in bounded regions in $\mathbb{R}^d$. Under the assumption that the discontinuity interface $Γ$ may be approximated by a connected series of hyperplanes with a prescribed accuracy $\varepsilon >0$, we show that a three-layer ReLU NN is sufficient to accurately approximate any piecewise constant function and establish its error bound. Moreover, if the discontinuity interface is convex, an analytical formula of the ReLU NN approximation with exact weights and biases is provided.
format Preprint
id arxiv_https___arxiv_org_abs_2410_16506
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle ReLU neural network approximation to piecewise constant functions
Cai, Zhiqiang
Choi, Junpyo
Liu, Min
Functional Analysis
Machine Learning
Numerical Analysis
68T07, 41A25, 41A46
This paper studies the approximation property of ReLU neural networks (NNs) to piecewise constant functions with unknown interfaces in bounded regions in $\mathbb{R}^d$. Under the assumption that the discontinuity interface $Γ$ may be approximated by a connected series of hyperplanes with a prescribed accuracy $\varepsilon >0$, we show that a three-layer ReLU NN is sufficient to accurately approximate any piecewise constant function and establish its error bound. Moreover, if the discontinuity interface is convex, an analytical formula of the ReLU NN approximation with exact weights and biases is provided.
title ReLU neural network approximation to piecewise constant functions
topic Functional Analysis
Machine Learning
Numerical Analysis
68T07, 41A25, 41A46
url https://arxiv.org/abs/2410.16506