Anderson-type acceleration method for Deep Neural Network optimization

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Ito, Kazufumi, Xue, Tiancheng
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866917136453074944
author Ito, Kazufumi
Xue, Tiancheng
author_facet Ito, Kazufumi
Xue, Tiancheng
contents In this paper we consider the neural network optimization. We develop Anderson-type acceleration method for the stochastic gradient decent method and it improves the network permanence very much. We demonstrate the applicability of the method for Deep Neural Network (DNN) and Convolution Neural Network (CNN).
format Preprint
id arxiv_https___arxiv_org_abs_2510_20254
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Anderson-type acceleration method for Deep Neural Network optimization
Ito, Kazufumi
Xue, Tiancheng
Numerical Analysis
Optimization and Control
In this paper we consider the neural network optimization. We develop Anderson-type acceleration method for the stochastic gradient decent method and it improves the network permanence very much. We demonstrate the applicability of the method for Deep Neural Network (DNN) and Convolution Neural Network (CNN).
title Anderson-type acceleration method for Deep Neural Network optimization
topic Numerical Analysis
Optimization and Control
url https://arxiv.org/abs/2510.20254