学校地图 | 网站地图 | ENGLISH
首页学校概况机构设置学科建设人才招聘招生就业数字校园校长信箱校友网
首页
 校园新闻 
 通知公告 
 学术交流 
 网站地图 
数学科学学院
当前位置: 首页>>学术交流>>数学科学学院>>正文
Second-Order Cone Reformulation Technique of Extended Trust-Region Problems
2017-11-09 09:46     (点击次数:)

报告题目:Second-Order Cone Reformulation Technique ofExtended Trust-Region Problems

报告人:北京邮电大学博士生导师艾文宝教授

时 间:2017年11月11日上午9:00(本周六上午9:00)

地 点:8号楼303(数学学院会议室)

摘 要:We consider the extended Celis-Dennis-Tapia (CDT)problem that has a positive duality gap. It is presented in theory that thispositive duality gap can be narrowed by adding an appropriate second-order-cone(SOC) constraint, which may lead to dividing the problem into two separatesubproblems. More concretely, for any extended CDT problem with a positive dualitygap, we prove that one SOC constraint is valid to narrow the positive dualitygap if and

only if the corresponding hyperplane intersects the openoptimal line segment. Especially when the second constraint function consistsof the product of two linear functions, we prove that the

positive duality gap can be eliminated thoroughly bysolving two subproblems with SOC constraints. For any classical CDT problemwith a positive duality gap, a new model with two SOC constraints

is proposed, and a sufficient condition is presentedunder which this positive duality gap can be eliminated thoroughly. Inparticular, based on the sufficient condition, it is proved that the positive

duality gaps of any two-dimensional classical CDTproblem and a class of three-dimensional classical CDT problems can beeliminated thoroughly. Numerical results of some gap-existing examples comingfrom other papers show that their positive duality gaps are indeed eliminatedby our SOC reformulation technique.

报告人简介:艾文宝,北京市教学名师,现任北京邮电大学数学系教授、博士生导师、数学系主任。主要研究方向为最优化理论与算法及其应用,在内点算法、非凸二次约束二次优化问题等领域取得了一系列重要的理论结果,这些结果发表在SIAM Optimation、Mathematical Programming等国际知名优化类顶尖学术期刊上。主持国家自然科学基金面上项目多项。目前是北京数学会理事、国际《数学评论》评论员,也是SIAM Optimation、MP等多个国际知名期刊的审稿人。

关闭窗口
 

版权所有 河南科技学院 河南省新乡市华兰大道东段 453003

豫公网安备 41070202000140号 工信部备案:豫ICP备05002482