Numerical block diagonalization of matrix *-algebras with application to semidefinite programming

E. de Klerk, C. Dobre, D.V. Pasechnik

Research output: Contribution to journalArticleScientificpeer-review

Abstract

Semidefinite programming (SDP) is one of the most active areas in mathematical programming, due to varied applications and the availability of interior point algorithms. In this paper we propose a new pre-processing technique for SDP instances that exhibit algebraic symmetry. We present computational results to show that the solution times of certain SDP instances may be greatly reduced via the new approach.
Original languageEnglish
Pages (from-to)91-111
JournalMathematical Programming
Volume129
Issue number1
Publication statusPublished - 2011

Fingerprint

Block Diagonalization
Mathematical programming
Semidefinite Programming
Matrix Algebra
Availability
Processing
Interior-point Algorithm
Mathematical Programming
Preprocessing
Computational Results
Symmetry

Cite this

@article{3270184874e844e6ba38279272afa067,
title = "Numerical block diagonalization of matrix *-algebras with application to semidefinite programming",
abstract = "Semidefinite programming (SDP) is one of the most active areas in mathematical programming, due to varied applications and the availability of interior point algorithms. In this paper we propose a new pre-processing technique for SDP instances that exhibit algebraic symmetry. We present computational results to show that the solution times of certain SDP instances may be greatly reduced via the new approach.",
author = "{de Klerk}, E. and C. Dobre and D.V. Pasechnik",
year = "2011",
language = "English",
volume = "129",
pages = "91--111",
journal = "Mathematical Programming",
issn = "0025-5610",
publisher = "Springer",
number = "1",

}

Numerical block diagonalization of matrix *-algebras with application to semidefinite programming. / de Klerk, E.; Dobre, C.; Pasechnik, D.V.

In: Mathematical Programming, Vol. 129, No. 1, 2011, p. 91-111.

Research output: Contribution to journalArticleScientificpeer-review

TY - JOUR

T1 - Numerical block diagonalization of matrix *-algebras with application to semidefinite programming

AU - de Klerk, E.

AU - Dobre, C.

AU - Pasechnik, D.V.

PY - 2011

Y1 - 2011

N2 - Semidefinite programming (SDP) is one of the most active areas in mathematical programming, due to varied applications and the availability of interior point algorithms. In this paper we propose a new pre-processing technique for SDP instances that exhibit algebraic symmetry. We present computational results to show that the solution times of certain SDP instances may be greatly reduced via the new approach.

AB - Semidefinite programming (SDP) is one of the most active areas in mathematical programming, due to varied applications and the availability of interior point algorithms. In this paper we propose a new pre-processing technique for SDP instances that exhibit algebraic symmetry. We present computational results to show that the solution times of certain SDP instances may be greatly reduced via the new approach.

M3 - Article

VL - 129

SP - 91

EP - 111

JO - Mathematical Programming

JF - Mathematical Programming

SN - 0025-5610

IS - 1

ER -