中文版 | English
题名

Variational Analysis Perspective on Linear Convergence of Some First Order Methods for Nonsmooth Convex Optimization Problems

作者
通讯作者Yuan,Xiaoming
发表日期
2021
DOI
发表期刊
ISSN
0927-6947
EISSN
1877-0541
卷号29页码:803-837
摘要

We study linear convergence of some first-order methods such as the proximal gradient method (PGM), the proximal alternating linearized minimization (PALM) algorithm and the randomized block coordinate proximal gradient method (R-BCPGM) for minimizing the sum of a smooth convex function and a nonsmooth convex function. We introduce a new analytic framework based on the error bound/calmness/metric subregularity/bounded metric subregularity. This variational analysis perspective enables us to provide some concrete sufficient conditions for linear convergence and applicable approaches for calculating linear convergence rates of these first-order methods for a class of structured convex problems. In particular, for the LASSO, the fused LASSO and the group LASSO, these conditions are satisfied automatically, and the modulus for the calmness/metric subregularity is computable. Consequently, the linear convergence of the first-order methods for these important applications is automatically guaranteed and the convergence rates can be calculated. The new perspective enables us to improve some existing results and obtain novel results unknown in the literature. Particularly, we improve the result on the linear convergence of the PGM and PALM for the structured convex problem with a computable error bound estimation. Also for the R-BCPGM for the structured convex problem, we prove that the linear convergence is ensured when the nonsmooth part of the objective function is the group LASSO regularizer.

关键词
相关链接[Scopus记录]
收录类别
SCI ; EI
语种
英语
学校署名
其他
WOS记录号
WOS:000668079100001
EI入藏号
20220811694483
Scopus记录号
2-s2.0-85109006557
来源库
Scopus
引用统计
被引频次[WOS]:13
成果类型期刊论文
条目标识符http://sustech.caswiz.com/handle/2SGJ60CL/242282
专题理学院_数学系
深圳国际数学中心(杰曼诺夫数学中心)(筹)
理学院_深圳国家应用数学中心
作者单位
1.Department of Mathematics and Statistics,University of Victoria,Victoria,V8P 5C2,Canada
2.Department of Mathematics,The University of Hong Kong,Hong Kong
3.Department of Mathematics,SUSTech International Center for Mathematics,Southern University of Science and Technology,National Center for Applied Mathematics Shenzhen,Shenzhen,China
推荐引用方式
GB/T 7714
Ye,Jane J.,Yuan,Xiaoming,Zeng,Shangzhi,et al. Variational Analysis Perspective on Linear Convergence of Some First Order Methods for Nonsmooth Convex Optimization Problems[J]. SET-VALUED ANAL,2021,29:803-837.
APA
Ye,Jane J.,Yuan,Xiaoming,Zeng,Shangzhi,&Zhang,Jin.(2021).Variational Analysis Perspective on Linear Convergence of Some First Order Methods for Nonsmooth Convex Optimization Problems.SET-VALUED ANAL,29,803-837.
MLA
Ye,Jane J.,et al."Variational Analysis Perspective on Linear Convergence of Some First Order Methods for Nonsmooth Convex Optimization Problems".SET-VALUED ANAL 29(2021):803-837.
条目包含的文件
条目无相关文件。
个性服务
原文链接
推荐该条目
保存到收藏夹
查看访问统计
导出为Endnote文件
导出为Excel格式
导出为Csv格式
Altmetrics Score
谷歌学术
谷歌学术中相似的文章
[Ye,Jane J.]的文章
[Yuan,Xiaoming]的文章
[Zeng,Shangzhi]的文章
百度学术
百度学术中相似的文章
[Ye,Jane J.]的文章
[Yuan,Xiaoming]的文章
[Zeng,Shangzhi]的文章
必应学术
必应学术中相似的文章
[Ye,Jane J.]的文章
[Yuan,Xiaoming]的文章
[Zeng,Shangzhi]的文章
相关权益政策
暂无数据
收藏/分享
所有评论 (0)
[发表评论/异议/意见]
暂无评论

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。