Guardado en:
Detalles Bibliográficos
Autores principales: Dereich, Steffen, Jentzen, Arnulf, Kassing, Sebastian
Formato: Preprint
Publicado: 2023
Materias:
Acceso en línea:https://arxiv.org/abs/2302.14690
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866915025367597056
author Dereich, Steffen
Jentzen, Arnulf
Kassing, Sebastian
author_facet Dereich, Steffen
Jentzen, Arnulf
Kassing, Sebastian
contents In this article, we show existence of minimizers in the loss landscape for residual artificial neural networks (ANNs) with multi-dimensional input layer and one hidden layer with ReLU activation. Our work contrasts earlier results in [D. Gallon, A. Jentzen, and F. Lindner, preprint, arXiv:2211.15641, 2022] and [P. Petersen, M. Raslan, and F. Voigtlaender, Found. Comput. Math., 21 (2021), pp. 375-444] which showed that in many situations minimizers do not exist for common smooth activation functions even in the case where the target functions are polynomials. The proof of the existence property makes use of a closure of the search space containing all functions generated by ANNs and additional discontinuous generalized responses. As we will show, the additional generalized responses in this larger space are suboptimal so that the minimum is attained in the original function class.
format Preprint
id arxiv_https___arxiv_org_abs_2302_14690
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle On the existence of minimizers in shallow residual ReLU neural network optimization landscapes
Dereich, Steffen
Jentzen, Arnulf
Kassing, Sebastian
Optimization and Control
Machine Learning
Numerical Analysis
Primary 68T07, Secondary 68T05, 41A50
In this article, we show existence of minimizers in the loss landscape for residual artificial neural networks (ANNs) with multi-dimensional input layer and one hidden layer with ReLU activation. Our work contrasts earlier results in [D. Gallon, A. Jentzen, and F. Lindner, preprint, arXiv:2211.15641, 2022] and [P. Petersen, M. Raslan, and F. Voigtlaender, Found. Comput. Math., 21 (2021), pp. 375-444] which showed that in many situations minimizers do not exist for common smooth activation functions even in the case where the target functions are polynomials. The proof of the existence property makes use of a closure of the search space containing all functions generated by ANNs and additional discontinuous generalized responses. As we will show, the additional generalized responses in this larger space are suboptimal so that the minimum is attained in the original function class.
title On the existence of minimizers in shallow residual ReLU neural network optimization landscapes
topic Optimization and Control
Machine Learning
Numerical Analysis
Primary 68T07, Secondary 68T05, 41A50
url https://arxiv.org/abs/2302.14690