维普资讯
发表评论我要收藏点击“我要推荐”按钮复制地址,将本页推荐给别人看,自己就可以获得积分奖励!点击“我要推荐”按钮复制地址,推荐文章给别人看,自己就可以获得积分奖励。

求解结构型单调变分不等式的改进的邻近类分解方法

下载全文
[全文大小:319 K]

李敏[1] 袁晓明[2]

[1]东南大学经济管理学院管理科学与工程系,南京210096 [2]上海交通大学,安泰经济与管理学院管理科学系,上海200052

应用数学和力学
订阅本刊
国际标准刊号:ISSN 1000-0887
国内统一刊号:CN 50-1060

摘  要:

邻近类分解方法首先是由Chen和Teboulle(Math.Programming,1994,64(1):81-101)提出用来求解凸的极小化问题.在此基础上,该文提出一种新方法求解具有分离结构的单调变分不等式.其主要优点在于放松了算法中对某些参数的限制,使得新方法更加便于计算.在和原分解方法相同的假设下,可以证明新方法是全局收敛的.[著者文摘]

Applied Mathematics and Mechanics

分 类 号:

O221 O224

文献标识码:

A

文章编号:

1000-0887(2007)12-1483-10

相关文章:

参考文献(18篇)  主题相关

[参考文献]

Improved Proximal-Based Decomposition Method for Structured Monotone Variational Inequalities

LI Min, YUAN Xiao-ming ( 1. Department of Management Science and Engineering, School of Economics and Management, Southeast University, Nanjing 210096, P. R. China; 2. Department of Management Science, Antai College of Economics and Management, Shanghai Jiaotong University, Shanghai 200052, P. R. China)

Abstract:

The proximal-based decomposition method was originally proposed by Chen and TebouUe (Math. Programming, 1994,64( 1 ) :81-101 ) for solving convex minimization problems. This paper extended to solve monotone variational inequalities associated with separable structures with the improvements that the restrictive assumptions on the involved parameters are much relaxed, and thus makes it practical to solve the involved subproblems easily. Without additional assumptions, global convergence of the new method is proved under the same mild assumptions on the problem' s data as the original method.[著者文摘]

Key words:

decomposition; inexact criterion; proximal; structured variational inequality

收稿日期: 2006-05-22
修订日期: 2007-09-14

基金资助:

国家自然科学基金资助项目(70671024);国家高新技术发展(863)计划(2006AA11Z209)

作者简介:

李敏(1980-),女,江苏徐州人,博士; 袁晓明,博士(联系人.Tel:+86-21—52301397;E—mail:xmyuan@sjtu.edu.cn).

更多评论>>文章评论
你是匿名用户 登录 | 注册 验证码 刷新
中国业务群个人门户,免费下载!
更多>>相关文章
天元数据 维普资讯 版权所有 Copyright © 2001-2008 cqvip.com Inc. All rights reserved.
渝ICP证 B2-20050021  违法和不良信息举报中心
建议使用:1024x768分辨率,16位以上颜色