摘要: |
通过引入新型邻近点参数修正策略及搜索方向子问题,提出一个求解非光滑优化的强次可行方向邻近点束方法.该方法稳定性好,能保证迭代点的强次可行性,且具备全局收敛性. |
关键词: 非光滑优化 邻近点 束方法 强次可行方向 |
DOI:10.13656/j.cnki.gxkx.20140328.001 |
投稿时间:2013-09-30修订日期:2013-12-11 |
基金项目:国家自然科学基金(11301095,11126341)和广西自然科学基金项目(2013GXNSFAA019013)资助。 |
|
A Proximal Bundle Method of Strongly Sub-feasible Directions for Nonsmooth Optimization |
TANG Chun-ming1, JIAN Jin-bao1,2
|
(1.College of Mathematics and Information Science, Guangxi University, Nanning, Guangxi, 530004, China;2.College of Mathematics and Information Science, Yulin Normal University, Yulin, Guangxi, 537000, China) |
Abstract: |
By introducing a new updating strategy of proximal parameters and a new search direction finding subproblem, a proximal bundle method of strongly sub-feasible directions is proposed for solving nonsmooth optimization.The proposed method is stable, can preserve sub-feasibility of the iterations, and possesses global convergence. |
Key words: nonsmooth optimization proximal point bundle method strongly sub-feasible directions |