Primal-dual interior-point methods for semidefinite programming: Convergence rates, stability and numerical results

被引:295
|
作者
Alizadeh, F [1 ]
Haeberly, JPA
Overton, ML
机构
[1] Rutgers State Univ, RUTCOR, New Brunswick, NJ 08854 USA
[2] Fordham Univ, Dept Math, Bronx, NY 10458 USA
[3] NYU, Courant Inst Math Sci, Dept Comp Sci, New York, NY 10012 USA
关键词
semidefinite programming; eigenvalue optimization; interior-point method; convex programming;
D O I
10.1137/S1052623496304700
中图分类号
O29 [应用数学];
学科分类号
070104 ;
摘要
Primal-dual interior-point path-following methods for semidefinite programming are considered. Several variants are discussed, based on Newton's method applied to three equations: primal feasibility, dual feasibility, and some form of centering condition. The focus is on three such algorithms, called the XZ, XZ+ZX, and Q methods. For the XZ+ZX and Q algorithms, the Newton system is well defined and its Jacobian is nonsingular at the solution, under nondegeneracy assumptions. The associated Schur complement matrix has an unbounded condition number on the central path under the nondegeneracy assumptions and an additional rank assumption. Practical aspects are discussed, including Mehrotra predictor-corrector variants and issues of numerical stability. Compared to the other methods considered, the XZ+ZX method is more robust with respect to its ability to step close to the boundary, converges more rapidly, and achieves higher accuracy.
引用
收藏
页码:746 / 768
页数:23
相关论文
共 50 条
  • [41] An efficient primal-dual interior-point method for minimizing a sum of Euclidean norms
    Andersen, KD
    Christiansen, E
    Conn, AR
    Overton, ML
    SIAM JOURNAL ON SCIENTIFIC COMPUTING, 2000, 22 (01) : 243 - 262
  • [42] Sensitivity analysis in linear programming and semidefinite programming using interior-point methods
    Yildirim, EA
    Todd, MJ
    MATHEMATICAL PROGRAMMING, 2001, 90 (02) : 229 - 261
  • [43] On homogeneous interior-point algorithms for semidefinite programming
    Potra, FA
    Sheng, RQ
    OPTIMIZATION METHODS & SOFTWARE, 1998, 9 (1-3) : 161 - 184
  • [44] A Primal-dual Interior-point Algorithm for Symmetric Cone Convex Quadratic Programming Based on the Commutative Class Directions
    S. Asadi
    H. Mansouri
    M. Zangiabadi
    Acta Mathematicae Applicatae Sinica, English Series, 2019, 35 : 359 - 373
  • [45] A Primal-dual Interior-point Algorithm for Symmetric Cone Convex Quadratic Programming Based on the Commutative Class Directions
    S.ASADI
    H.MANSOURI
    M.ZANGIABADI
    ActaMathematicaeApplicataeSinica, 2019, 35 (02) : 359 - 373
  • [46] A Primal-dual Interior-point Algorithm for Symmetric Cone Convex Quadratic Programming Based on the Commutative Class Directions
    Asadi, S.
    Mansouri, H.
    Zangiabadi, M.
    ACTA MATHEMATICAE APPLICATAE SINICA-ENGLISH SERIES, 2019, 35 (02): : 359 - 373
  • [47] Primal-dual interior-point algorithm based on a new kernel function for linear optimization
    Qian, Zhonggen
    Wang, Guoqiang
    Bai, Yanqin
    PROCEEDINGS OF THE SIXTH INTERNATIONAL CONFERENCE ON INFORMATION AND MANAGEMENT SCIENCES, 2007, 6 : 464 - 470
  • [48] A PRIMAL-DUAL INTERIOR-POINT METHOD CAPABLE OF RAPIDLY DETECTING INFEASIBILITY FOR NONLINEAR PROGRAMS
    Dai, Yu-Hong
    Liu, Xin-Wei
    Sun, Jie
    JOURNAL OF INDUSTRIAL AND MANAGEMENT OPTIMIZATION, 2020, 16 (02) : 1009 - 1035
  • [49] A comparative study of kernel functions for primal-dual interior-point algorithms in linear optimization
    Bai, YQ
    El Ghami, M
    Roos, C
    SIAM JOURNAL ON OPTIMIZATION, 2004, 15 (01) : 101 - 128
  • [50] Primal-dual potential reduction methods for semidefinite programming using affine-scaling directions
    de Klerk, E
    Roos, C
    Terlaky, T
    APPLIED NUMERICAL MATHEMATICS, 1999, 29 (03) : 335 - 360