基于近似离散梯度的梯度下降算法

2022 26th International Conference on System Theory, Control and Computing (ICSTCC) Pub Date : 2022-10-19 DOI:10.1109/ICSTCC55426.2022.9931872

Alessio Moreschini, Mattia Mattioni, S. Monaco, D. Normand-Cyrot

{"title":"基于近似离散梯度的梯度下降算法","authors":"Alessio Moreschini, Mattia Mattioni, S. Monaco, D. Normand-Cyrot","doi":"10.1109/ICSTCC55426.2022.9931872","DOIUrl":null,"url":null,"abstract":"We propose an optimization method obtained by the approximation of a novel discretization approach for gradient dynamics recently proposed by the authors. It is shown that the proposed algorithm ensures convergence for all amplitudes of the step size, contrarily to classical implementations.","PeriodicalId":220845,"journal":{"name":"2022 26th International Conference on System Theory, Control and Computing (ICSTCC)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2022-10-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"A gradient descent algorithm built on approximate discrete gradients\",\"authors\":\"Alessio Moreschini, Mattia Mattioni, S. Monaco, D. Normand-Cyrot\",\"doi\":\"10.1109/ICSTCC55426.2022.9931872\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"We propose an optimization method obtained by the approximation of a novel discretization approach for gradient dynamics recently proposed by the authors. It is shown that the proposed algorithm ensures convergence for all amplitudes of the step size, contrarily to classical implementations.\",\"PeriodicalId\":220845,\"journal\":{\"name\":\"2022 26th International Conference on System Theory, Control and Computing (ICSTCC)\",\"volume\":\"25 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2022-10-19\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2022 26th International Conference on System Theory, Control and Computing (ICSTCC)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ICSTCC55426.2022.9931872\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2022 26th International Conference on System Theory, Control and Computing (ICSTCC)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICSTCC55426.2022.9931872","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 0

摘要

我们提出了一种优化方法，该方法是由作者最近提出的一种新的梯度动力学离散化方法近似得到的。结果表明，与传统算法不同，该算法对所有步长幅值都具有收敛性。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

A gradient descent algorithm built on approximate discrete gradients

We propose an optimization method obtained by the approximation of a novel discretization approach for gradient dynamics recently proposed by the authors. It is shown that the proposed algorithm ensures convergence for all amplitudes of the step size, contrarily to classical implementations.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

2022 26th International Conference on System Theory, Control and Computing (ICSTCC)

自引率

0.00%

发文量