A New Wide Neighborhood Primal-Dual Infeasible-Interior-Point Method for Symmetric Cone Programming

被引:44
|
作者
Liu, Hongwei [1 ]
Yang, Ximei [1 ]
Liu, Changhe [2 ]
机构
[1] Xidian Univ, Dept Math, Xian 710071, Peoples R China
[2] Henan Univ Sci & Technol, Dept Appl Math, Luoyang 471003, Peoples R China
基金
中国国家自然科学基金;
关键词
Jordan algebra; Symmetric cone programming; Wide neighborhood; Infeasible-interior-point method; Polynomial complexity; POLYNOMIAL CONVERGENCE; ALGORITHMS;
D O I
10.1007/s10957-013-0303-y
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
We present a new infeasible-interior-point method, based on a wide neighborhood, for symmetric cone programming. The convergence is shown for a commutative class of search directions, which includes the Nesterov-Todd direction and the xs and sx directions. Moreover, we derive the complexity bound of the wide neighborhood infeasible interior-point methods that coincides with the currently best known theoretical complexity bounds for the short step path-following algorithm.
引用
收藏
页码:796 / 815
页数:20
相关论文
共 50 条