On duality theory for non-convex semidefinite programming

被引:3
|
作者
Sun, Wenyu [1 ]
Li, Chengjin [1 ,2 ]
Sampaio, Raimundo J. B. [3 ]
机构
[1] Nanjing Normal Univ, Jiangsu Key Lab NSLSCS, Sch Math Sci, Nanjing 210046, Peoples R China
[2] Fujian Normal Univ, Sch Math & Comp Sci, Fuzhou 350007, Peoples R China
[3] Pontif Catholic Univ Panara PUCPR, Grad Program Prod & Syst Engn PPGEPS, BR-81611970 Curitiba, Parana, Brazil
基金
中国国家自然科学基金;
关键词
Semidefinite programming; Nonconvex semidefinite programming; Duality; Convex-like function; Invex function; SUCCESSIVE LINEARIZATION METHODS;
D O I
10.1007/s10479-011-0861-z
中图分类号
C93 [管理学]; O22 [运筹学];
学科分类号
070105 ; 12 ; 1201 ; 1202 ; 120202 ;
摘要
In this paper, with the help of convex-like function, we discuss the duality theory for nonconvex semidefinite programming. Our contributions are: duality theory for the general nonconvex semidefinite programming when Slater's condition holds; perfect duality for a special case of the nonconvex semidefinite programming for which Slater's condition fails. We point out that the results of Fan (Appl. Math. Lett. 18:1068-1073, 2005) can be regarded as a special case of our result.
引用
收藏
页码:331 / 343
页数:13
相关论文
共 50 条