Newton methods for nonsmooth convex minimization: connections among U-Lagrangian, Riemannian Newton and SQP methods

Abstract : This paper studies Newton-type methods for minimization of partly smooth convex functions. Sequential Newton methods are provided using local parameterizations obtained from U-Lagrangian theory and from Riemannian geometry. The Hessian based on the U-Lagrangian depends on the selection of a dual parameter g; by revealing the connection to Riemannian geometry, a natural choice of g emerges for which the two Newton directions become identical. This choice of g is also shown to be related to the least-squares multiplier estimate from a sequential quadratic programming (SQP) approach, and with this multiplier, SQP gives the same search direction as the Newton methods.
Type de document :
Article dans une revue
Mathematical Programming B, Springer, 2005, 104, 〈10.1007/s10107-005-0631-2〉
Liste complète des métadonnées

Littérature citée [16 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/inria-00071403
Contributeur : Jérôme Malick <>
Soumis le : lundi 25 mars 2013 - 14:25:26
Dernière modification le : mercredi 11 avril 2018 - 01:52:24
Document(s) archivé(s) le : mercredi 26 juin 2013 - 04:01:31

Fichier

MM05.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

Collections

Citation

Scott A. Miller, Jérôme Malick. Newton methods for nonsmooth convex minimization: connections among U-Lagrangian, Riemannian Newton and SQP methods. Mathematical Programming B, Springer, 2005, 104, 〈10.1007/s10107-005-0631-2〉. 〈inria-00071403v2〉

Partager

Métriques

Consultations de la notice

179

Téléchargements de fichiers

220