Quasi-Newton acceleration for equality-constrained minimization

dc.creatorFerreira-Mendonca, L
dc.creatorLopes, VLR
dc.creatorMartinez, JM
dc.date2008
dc.dateJUL
dc.date2014-07-30T14:19:10Z
dc.date2015-11-26T17:38:21Z
dc.date2014-07-30T14:19:10Z
dc.date2015-11-26T17:38:21Z
dc.date.accessioned2018-03-29T00:19:58Z
dc.date.available2018-03-29T00:19:58Z
dc.identifierComputational Optimization And Applications. Springer, v. 40, n. 3, n. 373, n. 388, 2008.
dc.identifier0926-6003
dc.identifierWOS:000256754600004
dc.identifier10.1007/s10589-007-9090-4
dc.identifierhttp://www.repositorio.unicamp.br/jspui/handle/REPOSIP/58732
dc.identifierhttp://repositorio.unicamp.br/jspui/handle/REPOSIP/58732
dc.identifier.urihttp://repositorioslatinoamericanos.uchile.cl/handle/2250/1286181
dc.descriptionOptimality (or KKT) systems arise as primal-dual stationarity conditions for constrained optimization problems. Under suitable constraint qualifications, local minimizers satisfy KKT equations but, unfortunately, many other stationary points (including, perhaps, maximizers) may solve these nonlinear systems too. For this reason, nonlinear-programming solvers make strong use of the minimization structure and the naive use of nonlinear-system solvers in optimization may lead to spurious solutions. Nevertheless, in the basin of attraction of a minimizer, nonlinear-system solvers may be quite efficient. In this paper quasi-Newton methods for solving nonlinear systems are used as accelerators of nonlinear-programming (augmented Lagrangian) algorithms, with equality constraints. A periodically-restarted memoryless symmetric rank-one (SR1) correction method is introduced for that purpose. Convergence results are given and numerical experiments that confirm that the acceleration is effective are presented.
dc.description40
dc.description3
dc.description373
dc.description388
dc.languageen
dc.publisherSpringer
dc.publisherNew York
dc.publisherEUA
dc.relationComputational Optimization And Applications
dc.relationComput. Optim. Appl.
dc.rightsfechado
dc.rightshttp://www.springer.com/open+access/authors+rights?SGWID=0-176704-12-683201-0
dc.sourceWeb of Science
dc.subjectoptimality systems
dc.subjectquasi-Newton methods
dc.subjectminimization with equality constraints
dc.subjectSolving Nonlinear-systems
dc.subjectAugmented Lagrangian Algorithm
dc.subjectLinear-dependence Condition
dc.subjectProjected Gradient Methods
dc.subjectSimple Bounds
dc.subjectConvex-sets
dc.subjectOptimization
dc.subjectEquations
dc.subjectConvergence
dc.subjectQualification
dc.titleQuasi-Newton acceleration for equality-constrained minimization
dc.typeArtículos de revistas


Este ítem pertenece a la siguiente institución