Minimizing the sum of linear fractional functions over the cone of positive semidefinite matrices: Approximation and applications

被引：9

作者：

Xia, Yong ^{[1
]}

Wang, Longfei ^{[1
]}

Wang, Shu ^{[2
]}

机构：

[1] Beihang Univ, Sch Math & Syst Sci, State Key Lab Software Dev Environm, LMIB,Minist Educ, Beijing 100191, Peoples R China

[2] North China Inst Sci & Technol, Coll Sci, Langfang 065201, Hebei, Peoples R China

来源：

OPERATIONS RESEARCH LETTERS | 2018年 / 46卷 / 01期

关键词：

Fractional programming; Semidefinite programming; Rayleigh quotient; Total least squares; FPTAS; RATIOS PROBLEM; COMPLEXITY;

D O I：

10.1016/j.orl.2017.11.010

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：

The problem of maximizing the sum of two generalized Rayleigh quotients and the total least squares problem with nonsingular Tikhonov regularization are reformulated as a class of sum-of-linear-ratios minimizing over the cone of symmetric positive semidefinite matrices, which is shown to have a Fully Polynomial Time Approximation Scheme. (C) 2017 Elsevier B.V. All rights reserved.

引用

页码：76 / 80

页数：5

共 24 条

[1] Copositivity and constrained fractional quadratic problems [J].