A parallel inexact Newton method for stochastic programs with recourse

Xiaojun Chen; Robert S. Womersley

doi:10.1007/BF02187643

A parallel inexact Newton method for stochastic programs with recourse

Xiaojun Chen, Robert S. Womersley

Research output: Journal article publication › Journal article › Academic research › peer-review

14 Citations (Scopus)

Abstract

A parallel inexact Newton method with a line search is proposed for two-stage quadratic stochastic programs with recourse. A lattice rule is used for the numerical evaluation of multi-dimensional integrals, and a parallel iterative method is used to solve the quadratic programming subproblems. Although the objective only has a locally Lipschitz gradient, global convergence and local superlinear convergence of the method are established. Furthermore, the method provides an error estimate which does not require much extra computation. The performance of the method is illustrated on a CM5 parallel computer. Baltzer AG, Science Publishers.

Original language	English
Pages (from-to)	113-141
Number of pages	29
Journal	Annals of Operations Research
Volume	64
DOIs	https://doi.org/10.1007/BF02187643
Publication status	Published - 1 Jan 1996
Externally published	Yes

Keywords

Inexact newton method
Numerical integration
Parallel quadratic programming
Stochastic programming

ASJC Scopus subject areas

General Decision Sciences
Management Science and Operations Research

More information

10.1007/BF02187643

Cite this

@article{0749efc0443d47e3b37dc9eb63678aca,

title = "A parallel inexact Newton method for stochastic programs with recourse",

abstract = "A parallel inexact Newton method with a line search is proposed for two-stage quadratic stochastic programs with recourse. A lattice rule is used for the numerical evaluation of multi-dimensional integrals, and a parallel iterative method is used to solve the quadratic programming subproblems. Although the objective only has a locally Lipschitz gradient, global convergence and local superlinear convergence of the method are established. Furthermore, the method provides an error estimate which does not require much extra computation. The performance of the method is illustrated on a CM5 parallel computer. Baltzer AG, Science Publishers.",

keywords = "Inexact newton method, Numerical integration, Parallel quadratic programming, Stochastic programming",

author = "Xiaojun Chen and Womersley, {Robert S.}",

year = "1996",

month = jan,

day = "1",

doi = "10.1007/BF02187643",

language = "English",

volume = "64",

pages = "113--141",

journal = "Annals of Operations Research",

issn = "0254-5330",

publisher = "Springer Netherlands",

}

TY - JOUR

T1 - A parallel inexact Newton method for stochastic programs with recourse

AU - Chen, Xiaojun

AU - Womersley, Robert S.

PY - 1996/1/1

Y1 - 1996/1/1

N2 - A parallel inexact Newton method with a line search is proposed for two-stage quadratic stochastic programs with recourse. A lattice rule is used for the numerical evaluation of multi-dimensional integrals, and a parallel iterative method is used to solve the quadratic programming subproblems. Although the objective only has a locally Lipschitz gradient, global convergence and local superlinear convergence of the method are established. Furthermore, the method provides an error estimate which does not require much extra computation. The performance of the method is illustrated on a CM5 parallel computer. Baltzer AG, Science Publishers.

AB - A parallel inexact Newton method with a line search is proposed for two-stage quadratic stochastic programs with recourse. A lattice rule is used for the numerical evaluation of multi-dimensional integrals, and a parallel iterative method is used to solve the quadratic programming subproblems. Although the objective only has a locally Lipschitz gradient, global convergence and local superlinear convergence of the method are established. Furthermore, the method provides an error estimate which does not require much extra computation. The performance of the method is illustrated on a CM5 parallel computer. Baltzer AG, Science Publishers.

KW - Inexact newton method

KW - Numerical integration

KW - Parallel quadratic programming

KW - Stochastic programming

UR - http://www.scopus.com/inward/record.url?scp=21344434512&partnerID=8YFLogxK

U2 - 10.1007/BF02187643

DO - 10.1007/BF02187643

M3 - Journal article

SN - 0254-5330

VL - 64

SP - 113

EP - 141

JO - Annals of Operations Research

JF - Annals of Operations Research

ER -

A parallel inexact Newton method for stochastic programs with recourse

Abstract

Keywords

ASJC Scopus subject areas

More information

Other files and links

Fingerprint

Cite this