Semidefinite diagonal directions Monte Carlo algorithms for detecting necessary linear matrix inequality constraints

Shafiu Jibrin, Arnon Boneh, Jackie Van Ryzin

Research output: Contribution to journalArticle

Abstract

Hit-and-run algorithms are Monte Carlo methods for detecting necessary constraints in convex programming including semidefinite programming. The well known of these in semidefinite programming are semidefinite coordinate directions (SCD), semidefinite hypersphere directions (SHD) and semidefinite stand-and-hit (SSH) algorithms. SCD is considered to be the best on average and hence we use it for comparison. We develop two new hit-and-run algorithms in semidefinite programming that use diagonal directions. They are the uniform semidefinite diagonal directions (uniform SDD) and the original semidefinite diagonal directions (original SDD) algorithms. We analyze the costs and benefits of this change in comparison with SCD. We also show that both uniform SDD and original SDD generate points that are asymptotically uniform in the interior of the feasible region defined by the constraints.

Original languageEnglish (US)
Pages (from-to)2277-2288
Number of pages12
JournalCommunications in Nonlinear Science and Numerical Simulation
Volume14
Issue number5
DOIs
StatePublished - May 2009

Fingerprint

Monte Carlo Algorithm
Inequality Constraints
Linear matrix inequalities
Matrix Inequality
Linear Inequalities
Necessary
Semidefinite Programming
Hits
Convex optimization
Monte Carlo methods
Hypersphere
Feasible region
Convex Programming
Costs
Monte Carlo method
Interior

Keywords

  • Linear matrix inequalities
  • Monte Carlo method
  • Redundancy
  • Semidefinite programming

ASJC Scopus subject areas

  • Modeling and Simulation
  • Numerical Analysis
  • Applied Mathematics

Cite this

Semidefinite diagonal directions Monte Carlo algorithms for detecting necessary linear matrix inequality constraints. / Jibrin, Shafiu; Boneh, Arnon; Van Ryzin, Jackie.

In: Communications in Nonlinear Science and Numerical Simulation, Vol. 14, No. 5, 05.2009, p. 2277-2288.

Research output: Contribution to journalArticle

@article{8daf981594644160a585073f564c7ecf,
title = "Semidefinite diagonal directions Monte Carlo algorithms for detecting necessary linear matrix inequality constraints",
abstract = "Hit-and-run algorithms are Monte Carlo methods for detecting necessary constraints in convex programming including semidefinite programming. The well known of these in semidefinite programming are semidefinite coordinate directions (SCD), semidefinite hypersphere directions (SHD) and semidefinite stand-and-hit (SSH) algorithms. SCD is considered to be the best on average and hence we use it for comparison. We develop two new hit-and-run algorithms in semidefinite programming that use diagonal directions. They are the uniform semidefinite diagonal directions (uniform SDD) and the original semidefinite diagonal directions (original SDD) algorithms. We analyze the costs and benefits of this change in comparison with SCD. We also show that both uniform SDD and original SDD generate points that are asymptotically uniform in the interior of the feasible region defined by the constraints.",
keywords = "Linear matrix inequalities, Monte Carlo method, Redundancy, Semidefinite programming",
author = "Shafiu Jibrin and Arnon Boneh and {Van Ryzin}, Jackie",
year = "2009",
month = "5",
doi = "10.1016/j.cnsns.2008.07.005",
language = "English (US)",
volume = "14",
pages = "2277--2288",
journal = "Communications in Nonlinear Science and Numerical Simulation",
issn = "1007-5704",
publisher = "Elsevier",
number = "5",

}

TY - JOUR

T1 - Semidefinite diagonal directions Monte Carlo algorithms for detecting necessary linear matrix inequality constraints

AU - Jibrin, Shafiu

AU - Boneh, Arnon

AU - Van Ryzin, Jackie

PY - 2009/5

Y1 - 2009/5

N2 - Hit-and-run algorithms are Monte Carlo methods for detecting necessary constraints in convex programming including semidefinite programming. The well known of these in semidefinite programming are semidefinite coordinate directions (SCD), semidefinite hypersphere directions (SHD) and semidefinite stand-and-hit (SSH) algorithms. SCD is considered to be the best on average and hence we use it for comparison. We develop two new hit-and-run algorithms in semidefinite programming that use diagonal directions. They are the uniform semidefinite diagonal directions (uniform SDD) and the original semidefinite diagonal directions (original SDD) algorithms. We analyze the costs and benefits of this change in comparison with SCD. We also show that both uniform SDD and original SDD generate points that are asymptotically uniform in the interior of the feasible region defined by the constraints.

AB - Hit-and-run algorithms are Monte Carlo methods for detecting necessary constraints in convex programming including semidefinite programming. The well known of these in semidefinite programming are semidefinite coordinate directions (SCD), semidefinite hypersphere directions (SHD) and semidefinite stand-and-hit (SSH) algorithms. SCD is considered to be the best on average and hence we use it for comparison. We develop two new hit-and-run algorithms in semidefinite programming that use diagonal directions. They are the uniform semidefinite diagonal directions (uniform SDD) and the original semidefinite diagonal directions (original SDD) algorithms. We analyze the costs and benefits of this change in comparison with SCD. We also show that both uniform SDD and original SDD generate points that are asymptotically uniform in the interior of the feasible region defined by the constraints.

KW - Linear matrix inequalities

KW - Monte Carlo method

KW - Redundancy

KW - Semidefinite programming

UR - http://www.scopus.com/inward/record.url?scp=56049122682&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=56049122682&partnerID=8YFLogxK

U2 - 10.1016/j.cnsns.2008.07.005

DO - 10.1016/j.cnsns.2008.07.005

M3 - Article

VL - 14

SP - 2277

EP - 2288

JO - Communications in Nonlinear Science and Numerical Simulation

JF - Communications in Nonlinear Science and Numerical Simulation

SN - 1007-5704

IS - 5

ER -