Abstract
The limiting factors of second-order methods for large-scale semidefinite optimization are the storage and factorization of the Newton matrix. For a particular algorithm based on the modified barrier method, we propose to use iterative solvers instead of the routinely used direct factorization techniques. The preconditioned conjugate gradient method proves to be a viable alternative for problems with a large number of variables and modest size of the constrained matrix. We further propose to avoid explicit calculation of the Newton matrix either by an implicit scheme in the matrix–vector product or using a finite-difference formula. This leads to huge savings in memory requirements and, for certain problems, to further speed-up of the algorithm.
Similar content being viewed by others
References
Alizadeh F., Haeberly J.P.A., Overton M.L. (1998) Primal–dual interior–point methods for semidefinite programming: convergence rates, stability and numerical results. SIAM J. Optim. 8, 746–768
Benson S.J., Ye Y., Zhang X. (2000) Solving large-scale sparse semidefinite programs for combinatorial optimization. SIAM J. Optim. 10, 443–462
Borchers, B.: SDPLIB 1.2, a library of semidefinite programming test problems. Optim. Methods Soft. 11 & 12, 683–690 (1999). Available at http://www.nmt.edu/~borchers/
Burer S., Monteiro R. (2003) A nonlinear programming algorithm for solving semidefinite programs via low-rank factorization. Math. Prog. (Ser. B), 95(2): 329–357
Choi, C., Ye, Y.: Solving sparse semidefinite programs using the dual scaling algorithm with an iterative solver. In: Working Paper, Computational Optimization Laboratory, University of Iowa, Iowa City, IA (2000)
Dolan E.D., Moré J. (2002) Benchmarking optimization software with performance profiles. Math. Prog. 91, 201–213
Fujisawa K., Kojima M., Nakata K. (1997) Exploiting sparsity in primal-dual interior-point method for semidefinite programming. Math. Prog. 79, 235–253
Fukuda M., Kojima M., Shida M. (2002) Lagrangian dual interior-point methods for semidefinite programs. SIAM J. Optim. 12, 1007–1031
Geiger, C., Kanzow, C.: Numerische Verfahren zur Lösung unrestringierter Optimierungsaufgaben. Springer, Berlin Heidelberg New York (1999) (In German)
Helmberg C., Rendl F., Vanderbei R.J., Wolkowicz H. (1996) An interior–point method for semidefinite programming. SIAM J. Optim. 6, 342–361
Kočvara, M., Leibfritz, F., Stingl, M., Henrion, D.: A nonlinear SDP algorithm for static output feedback problems in COMPlib. LAAS-CNRS Research Report No. 04508, LAAS, Toulouse (2004)
Kočvara M., Stingl M. (2003) PENNON—a code for convex nonlinear and semidefinite programming. Optim. Methods Softw. 18, 317–333
Kočvara M., Stingl M. (2004) Solving nonconvex SDP problems of structural optimization with stability control. Optim. Methods Softw. 19, 595–609
Lin C.J., Saigal R. (2000) An incomplete cholesky factorization for dense matrices. BIT 40, 536–558
Mittelmann, H.: Benchmarks for Optimization Software; as of 26 November 2005. Available at http://plato.la.asu.edu/bench.html
Mittelmann H.D. (2003) An independent benchmarking of SDP and SOCP solvers. Math. Prog. 95, 407–430
Monteiro R.D.C. (1997) Primal–dual path-following algorithms for semidefinite programming. SIAM J. Optim. 7, 663–678
Morales J.L., Nocedal J. (2000) Automatic preconditioning by limited memory quasi-Newton updating. SIAM J. Optim. 10, 1079–1096
Nocedal J., Wright S. (1999) Numerical Optimization. Springer Series in Operations Research. Springer, Berlin Heidelberg New York
Polyak R. (1992) Modified barrier functions: Theory and methods. Math. Prog. 54, 177–222
Stingl, M.: On the solution of nonlinear semidefinite programs by augmented Lagrangian methods. Ph.D. Thesis, Institute of Applied Mathematics II, Friedrich-Alexander University of Erlangen-Nuremberg (2005) (submitted)
Sturm, J.: Primal-dual interior point approach to semidefinite programming. Ph.D. Thesis, Tinbergen Institute Research Series vol. 156, Thesis Publishers, Amsterdam, The Netherlands (1997). Available at http://members.tripodnet.nl/SeDuMi/sturm/papers/ thesisSTURM.ps.gz
Toh K.C. (2003) Solving large scale semidefinite programs via an iterative solver on the augmented systems. SIAM J. Optim. 14, 670–698
Toh K.C., Kojima M. (2002) Solving some large scale semidefinite programs via the conjugate residual method. SIAM J. Optim. 12, 669–691
Trick, M., Chvátal, V., Cook, W., Johnson, D., McGeoch, C., Trajan, R.: The second DIMACS implementation challenge: NP hard problems: Maximum clique, graph coloring, and satisfiability. Technical Report, Rutgers University. Available at http://dimacs.rutgers.edu/Challenges/
Zhang S.L., Nakata K., Kojima M. (2002) Incomplete orthogonalization preconditioners for solving large and dense linear systems which arise from semidefinite programming. Appl. Numer. Math. 41, 235–245
Author information
Authors and Affiliations
Corresponding author
Additional information
Dedicated to the memory of Jos Sturm.
An erratum to this article is available at http://dx.doi.org/10.1007/s10107-008-0250-9.
Rights and permissions
About this article
Cite this article
Kočvara, M., Stingl, M. On the solution of large-scale SDP problems by the modified barrier method using iterative solvers. Math. Program. 109, 413–444 (2007). https://doi.org/10.1007/s10107-006-0029-9
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10107-006-0029-9