大规模一般形式正则化的混合 LSMR 算法

Yanfei Yang
{"title":"大规模一般形式正则化的混合 LSMR 算法","authors":"Yanfei Yang","doi":"arxiv-2409.09104","DOIUrl":null,"url":null,"abstract":"The hybrid LSMR algorithm is proposed for large-scale general-form\nregularization. It is based on a Krylov subspace projection method where the\nmatrix $A$ is first projected onto a subspace, typically a Krylov subspace,\nwhich is implemented via the Golub-Kahan bidiagonalization process applied to\n$A$, with starting vector $b$. Then a regularization term is employed to the\nprojections. Finally, an iterative algorithm is exploited to solve a least\nsquares problem with constraints. The resulting algorithms are called the\n{hybrid LSMR algorithm}. At every step, we exploit LSQR algorithm to solve the\ninner least squares problem, which is proven to become better conditioned as\nthe number of $k$ increases, so that the LSQR algorithm converges faster. We\nprove how to select the stopping tolerances for LSQR in order to guarantee that\nthe regularized solution obtained by iteratively computing the inner least\nsquares problems and the one obtained by exactly computing the inner least\nsquares problems have the same accuracy. Numerical experiments illustrate that\nthe best regularized solution by the hybrid LSMR algorithm is as accurate as\nthat by JBDQR which is a joint bidiagonalization based algorithm.","PeriodicalId":501162,"journal":{"name":"arXiv - MATH - Numerical Analysis","volume":"101 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2024-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Hybrid LSMR algorithms for large-scale general-form regularization\",\"authors\":\"Yanfei Yang\",\"doi\":\"arxiv-2409.09104\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"The hybrid LSMR algorithm is proposed for large-scale general-form\\nregularization. It is based on a Krylov subspace projection method where the\\nmatrix $A$ is first projected onto a subspace, typically a Krylov subspace,\\nwhich is implemented via the Golub-Kahan bidiagonalization process applied to\\n$A$, with starting vector $b$. Then a regularization term is employed to the\\nprojections. Finally, an iterative algorithm is exploited to solve a least\\nsquares problem with constraints. The resulting algorithms are called the\\n{hybrid LSMR algorithm}. At every step, we exploit LSQR algorithm to solve the\\ninner least squares problem, which is proven to become better conditioned as\\nthe number of $k$ increases, so that the LSQR algorithm converges faster. We\\nprove how to select the stopping tolerances for LSQR in order to guarantee that\\nthe regularized solution obtained by iteratively computing the inner least\\nsquares problems and the one obtained by exactly computing the inner least\\nsquares problems have the same accuracy. Numerical experiments illustrate that\\nthe best regularized solution by the hybrid LSMR algorithm is as accurate as\\nthat by JBDQR which is a joint bidiagonalization based algorithm.\",\"PeriodicalId\":501162,\"journal\":{\"name\":\"arXiv - MATH - Numerical Analysis\",\"volume\":\"101 1\",\"pages\":\"\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-09-13\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - MATH - Numerical Analysis\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2409.09104\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - MATH - Numerical Analysis","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2409.09104","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

混合 LSMR 算法是针对大规模一般形式正规化而提出的。该算法基于 Krylov 子空间投影法,首先将矩阵 $A$ 投影到一个子空间,通常是 Krylov 子空间,该子空间是通过应用于 $A$ 的 Golub-Kahan 二对角化过程实现的,起始向量为 $b$。然后对投影采用正则化项。最后,利用迭代算法解决带有约束条件的最小二乘问题。由此产生的算法称为{混合 LSMR 算法}。在每一步中,我们都利用 LSQR 算法求解内最小二乘问题,事实证明,随着 $k$ 数量的增加,条件会变得更好,因此 LSQR 算法收敛得更快。我们探讨了如何选择 LSQR 的停止公差,以保证通过迭代计算内最小二乘问题得到的正则化解和通过精确计算内最小二乘问题得到的正则化解具有相同的精度。数值实验表明,混合 LSMR 算法的最佳正则化解与基于联合对角线算法的 JBDQR 算法的精度相同。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Hybrid LSMR algorithms for large-scale general-form regularization
The hybrid LSMR algorithm is proposed for large-scale general-form regularization. It is based on a Krylov subspace projection method where the matrix $A$ is first projected onto a subspace, typically a Krylov subspace, which is implemented via the Golub-Kahan bidiagonalization process applied to $A$, with starting vector $b$. Then a regularization term is employed to the projections. Finally, an iterative algorithm is exploited to solve a least squares problem with constraints. The resulting algorithms are called the {hybrid LSMR algorithm}. At every step, we exploit LSQR algorithm to solve the inner least squares problem, which is proven to become better conditioned as the number of $k$ increases, so that the LSQR algorithm converges faster. We prove how to select the stopping tolerances for LSQR in order to guarantee that the regularized solution obtained by iteratively computing the inner least squares problems and the one obtained by exactly computing the inner least squares problems have the same accuracy. Numerical experiments illustrate that the best regularized solution by the hybrid LSMR algorithm is as accurate as that by JBDQR which is a joint bidiagonalization based algorithm.
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
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信