A multidimensional filter algorithm for nonlinear equations and nonlinear least-squares

Nick Gould; Sven Leyffer; Philippe Toint

doi:10.1137/S1052623403422637

A multidimensional filter algorithm for nonlinear equations and nonlinear least-squares

Nick Gould, Sven Leyffer, Philippe Toint

Résultats de recherche: Contribution à un journal/une revue › Article › Revue par des pairs

46 Téléchargements (Pure)

Résumé

We introduce a new algorithm for the solution of systems of nonlinear equations and nonlinear least-squares problems that attempts to combine the efficiency of filter techniques and the robustness of trust-region methods. The algorithm is shown, under reasonable assumptions, to globally converge to zeros of the system, or to first-order stationary points of the Euclidean norm of its residual. Preliminary numerical experience is presented that shows substantial gains in efficiency over the traditional monotone trust-region approach. © 2004 Society for Industrial and Applied Mathematics.

langue originale	Anglais
Pages (de - à)	17-38
Nombre de pages	22
journal	SIAM Journal on Optimization
Volume	15
Numéro de publication	1
Les DOIs	https://doi.org/10.1137/S1052623403422637
Etat de la publication	Publié - 1 janv. 2005

SDG des Nations Unies

Ce résultat contribue à ou aux Objectifs de développement durable suivants

Accès au document

10.1137/S1052623403422637

45187Accepted author manuscript, 312 KB

Autres fichiers et liens

Link to publication in Scopus

1 Terminé
1 Actif

ADALGOPT: ADALGOPT - Algorithmes avancés en optimisation non-linéaire
Sartenaer, A. & Toint, P.
1/01/87 → …
Projet: Axe de recherche
Algorithmes de filtre pour les problèmes nonlinéaires
Toint, P. & SAINVITU, C.
2/02/02 → 31/08/11
Projet: Recherche

Contient cette citation

@article{d033a151b666442289e671e7e98c894d,

title = "A multidimensional filter algorithm for nonlinear equations and nonlinear least-squares",

abstract = "We introduce a new algorithm for the solution of systems of nonlinear equations and nonlinear least-squares problems that attempts to combine the efficiency of filter techniques and the robustness of trust-region methods. The algorithm is shown, under reasonable assumptions, to globally converge to zeros of the system, or to first-order stationary points of the Euclidean norm of its residual. Preliminary numerical experience is presented that shows substantial gains in efficiency over the traditional monotone trust-region approach. {\textcopyright} 2004 Society for Industrial and Applied Mathematics.",

keywords = " least-squares methods, filter methods, nonlinear equations, Nonlinear optimization, convergence theory",

author = "Nick Gould and Sven Leyffer and Philippe Toint",

year = "2005",

month = jan,

day = "1",

doi = "10.1137/S1052623403422637",

language = "English",

volume = "15",

pages = "17--38",

journal = "SIAM Journal on Optimization",

issn = "1095-7189",

publisher = "Society for Industrial and Applied Mathematics",

number = "1",

}

TY - JOUR

T1 - A multidimensional filter algorithm for nonlinear equations and nonlinear least-squares

AU - Gould, Nick

AU - Leyffer, Sven

AU - Toint, Philippe

PY - 2005/1/1

Y1 - 2005/1/1

N2 - We introduce a new algorithm for the solution of systems of nonlinear equations and nonlinear least-squares problems that attempts to combine the efficiency of filter techniques and the robustness of trust-region methods. The algorithm is shown, under reasonable assumptions, to globally converge to zeros of the system, or to first-order stationary points of the Euclidean norm of its residual. Preliminary numerical experience is presented that shows substantial gains in efficiency over the traditional monotone trust-region approach. © 2004 Society for Industrial and Applied Mathematics.

AB - We introduce a new algorithm for the solution of systems of nonlinear equations and nonlinear least-squares problems that attempts to combine the efficiency of filter techniques and the robustness of trust-region methods. The algorithm is shown, under reasonable assumptions, to globally converge to zeros of the system, or to first-order stationary points of the Euclidean norm of its residual. Preliminary numerical experience is presented that shows substantial gains in efficiency over the traditional monotone trust-region approach. © 2004 Society for Industrial and Applied Mathematics.

KW - least-squares methods

KW - filter methods

KW - nonlinear equations

KW - Nonlinear optimization

KW - convergence theory

UR - http://www.scopus.com/inward/record.url?scp=14944339092&partnerID=8YFLogxK

U2 - 10.1137/S1052623403422637

DO - 10.1137/S1052623403422637

M3 - Article

SN - 1095-7189

VL - 15

SP - 17

EP - 38

JO - SIAM Journal on Optimization

JF - SIAM Journal on Optimization

IS - 1

ER -

A multidimensional filter algorithm for nonlinear equations and nonlinear least-squares

Résumé

SDG des Nations Unies

Accès au document

Autres fichiers et liens

Empreinte digitale

Projets

ADALGOPT: ADALGOPT - Algorithmes avancés en optimisation non-linéaire

Algorithmes de filtre pour les problèmes nonlinéaires

Contient cette citation