Skip to Main content Skip to Navigation
Journal articles

Convergence of descent methods for semi-algebraic and tame problems: proximal algorithms, forward-backward splitting, and regularized Gauss-Seidel methods

Hedy Attouch 1 Jérôme Bolte 2 Benar Svaiter 3
2 TSE (GREMAQ, Université Toulouse I)
GREMAQ - Groupe de recherche en économie mathématique et quantitative
3 IMPA
IMPA - Instituto Nacional de Matemática Pura e Aplicada
Abstract : In view of the minimization of a nonsmooth nonconvex function f , we prove an abstract convergence result for descent methods satisfying a sufficientdecrease assumption, and allowing a relative error tolerance. Our result guarantees the convergence of bounded sequences, under the assumption that the function f satisfies the Kurdyka-Łojasiewicz inequality. This assumption allows to cover a wide range of problems, including nonsmooth semi-algebraic (or more generally tame) minimization. The specialization of our result to different kinds of structured problems provides several new convergence results for inexact versions of the gradient method, the proximal method, the forward-backward splitting algorithm, the gradient projection and some proximal regularization of the Gauss-Seidel method in a nonconvex setting. Our results are illustrated through feasibility problems, or iterative thresholding procedures for compressive sensing.
Document type :
Journal articles
Complete list of metadatas

Cited literature [57 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-00790042
Contributor : Hedy Attouch <>
Submitted on : Tuesday, February 19, 2013 - 12:09:07 PM
Last modification on : Friday, May 29, 2020 - 7:48:12 PM
Document(s) archivé(s) le : Monday, May 20, 2013 - 4:01:13 AM

File

ABS--MPA-Revised-21July2011.pd...
Files produced by the author(s)

Identifiers

Citation

Hedy Attouch, Jérôme Bolte, Benar Svaiter. Convergence of descent methods for semi-algebraic and tame problems: proximal algorithms, forward-backward splitting, and regularized Gauss-Seidel methods. Mathematical Programming, Series A, Springer, 2011, 137 (1), pp.91-124. ⟨10.1007/s10107-011-0484-9⟩. ⟨hal-00790042⟩

Share

Metrics

Record views

1225

Files downloads

1862