Enregistré dans:
Détails bibliographiques
Auteurs principaux: Alcantara, Jan Harold, Lee, Ching-pei
Format: Preprint
Publié: 2022
Sujets:
Accès en ligne:https://arxiv.org/abs/2211.02271
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866910020246962176
author Alcantara, Jan Harold
Lee, Ching-pei
author_facet Alcantara, Jan Harold
Lee, Ching-pei
contents We consider the projected gradient algorithm for the nonconvex best subset selection problem that minimizes a given empirical loss function under an $\ell_0$-norm constraint. Through decomposing the feasible set of the given sparsity constraint as a finite union of linear subspaces, we present two acceleration schemes with global convergence guarantees, one by same-space extrapolation and the other by subspace identification. The former fully utilizes the problem structure to greatly accelerate the optimization speed with only negligible additional cost. The latter leads to a two-stage meta-algorithm that first uses classical projected gradient iterations to identify the correct subspace containing an optimal solution, and then switches to a highly-efficient smooth optimization method in the identified subspace to attain superlinear convergence. Experiments demonstrate that the proposed accelerated algorithms are magnitudes faster than their non-accelerated counterparts as well as the state of the art.
format Preprint
id arxiv_https___arxiv_org_abs_2211_02271
institution arXiv
publishDate 2022
record_format arxiv
spellingShingle Accelerated projected gradient algorithms for sparsity constrained optimization problems
Alcantara, Jan Harold
Lee, Ching-pei
Optimization and Control
We consider the projected gradient algorithm for the nonconvex best subset selection problem that minimizes a given empirical loss function under an $\ell_0$-norm constraint. Through decomposing the feasible set of the given sparsity constraint as a finite union of linear subspaces, we present two acceleration schemes with global convergence guarantees, one by same-space extrapolation and the other by subspace identification. The former fully utilizes the problem structure to greatly accelerate the optimization speed with only negligible additional cost. The latter leads to a two-stage meta-algorithm that first uses classical projected gradient iterations to identify the correct subspace containing an optimal solution, and then switches to a highly-efficient smooth optimization method in the identified subspace to attain superlinear convergence. Experiments demonstrate that the proposed accelerated algorithms are magnitudes faster than their non-accelerated counterparts as well as the state of the art.
title Accelerated projected gradient algorithms for sparsity constrained optimization problems
topic Optimization and Control
url https://arxiv.org/abs/2211.02271