We propose an adaptive refinement algorithm to solve total variation regularized measure optimization problems. The method iteratively constructs dyadic partitions of the unit cube based on (i) the resolution of discretized dual problems and (ii) the detection of cells containing points that violate...

Full description

Saved in:
Bibliographic Details
Main Authors: Flinth, Axel, de Gournay, Frédéric, Weiss, Pierre
Format: Article
Language:English
Published: Université de Montpellier 2025-03-01
Series:Open Journal of Mathematical Optimization
Subjects:
Online Access:https://ojmo.centre-mersenne.org/articles/10.5802/ojmo.39/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849247390812340224
author Flinth, Axel
de Gournay, Frédéric
Weiss, Pierre
author_facet Flinth, Axel
de Gournay, Frédéric
Weiss, Pierre
author_sort Flinth, Axel
collection DOAJ
description We propose an adaptive refinement algorithm to solve total variation regularized measure optimization problems. The method iteratively constructs dyadic partitions of the unit cube based on (i) the resolution of discretized dual problems and (ii) the detection of cells containing points that violate the dual constraints. The detection is based on upper-bounds on the dual certificate, in the spirit of branch-and-bound methods. The interest of this approach is that it avoids the use of heuristic approaches to find the maximizers of dual certificates. We prove the convergence of this approach under mild hypotheses and a linear convergence rate under additional non-degeneracy assumptions. These results are confirmed by simple numerical experiments.
format Article
id doaj-art-011aca7180f04dc4b4e9efab87f48985
institution Kabale University
issn 2777-5860
language English
publishDate 2025-03-01
publisher Université de Montpellier
record_format Article
series Open Journal of Mathematical Optimization
spelling doaj-art-011aca7180f04dc4b4e9efab87f489852025-08-20T03:58:14ZengUniversité de MontpellierOpen Journal of Mathematical Optimization2777-58602025-03-01612710.5802/ojmo.3910.5802/ojmo.39Flinth, Axel0de Gournay, Frédéric1Weiss, Pierre2Department of Mathematics and Mathematical Statistics, Umeå University, SwedenInstitut de Mathématiques de Toulouse (IMT), Université de Toulouse, CNRS, INSA, FranceInstitut de Recherche en Informatique de Toulouse (IRIT), Université de Toulouse, CNRS, Centre de Biologie Intégrative (CBI), Laboratoire de biologie Moléculaire, Cellulaire et Développement (MCD), FranceWe propose an adaptive refinement algorithm to solve total variation regularized measure optimization problems. The method iteratively constructs dyadic partitions of the unit cube based on (i) the resolution of discretized dual problems and (ii) the detection of cells containing points that violate the dual constraints. The detection is based on upper-bounds on the dual certificate, in the spirit of branch-and-bound methods. The interest of this approach is that it avoids the use of heuristic approaches to find the maximizers of dual certificates. We prove the convergence of this approach under mild hypotheses and a linear convergence rate under additional non-degeneracy assumptions. These results are confirmed by simple numerical experiments.https://ojmo.centre-mersenne.org/articles/10.5802/ojmo.39/Total variationmeasure spacesFrank–Wolfe
spellingShingle Flinth, Axel
de Gournay, Frédéric
Weiss, Pierre
Open Journal of Mathematical Optimization
Total variation
measure spaces
Frank–Wolfe
topic Total variation
measure spaces
Frank–Wolfe
url https://ojmo.centre-mersenne.org/articles/10.5802/ojmo.39/