Coordinate Descent for SLOPE - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2023

Coordinate Descent for SLOPE

Johan Larsson
  • Fonction : Auteur
  • PersonId : 1285491
Quentin Klopfenstein
  • Fonction : Auteur
  • PersonId : 1285492
Jonas Wallin
  • Fonction : Auteur
  • PersonId : 1285493

Résumé

The lasso is the most famous sparse regression and feature selection method. One reason for its popularity is the speed at which the underlying optimization problem can be solved. Sorted LOne Penalized Estimation (SLOPE) is a generalization of the lasso with appealing statistical properties. In spite of this, the method has not yet reached widespread interest. A major reason for this is that current software packages that fit SLOPE rely on algorithms that perform poorly in high dimensions. To tackle this issue, we propose a new fast algorithm to solve the SLOPE optimization problem, which combines proximal gradient descent and proximal coordinate descent steps. We provide new results on the directional derivative of the SLOPE penalty and its related SLOPE thresholding operator, as well as provide convergence guarantees for our proposed solver. In extensive benchmarks on simulated and real data, we demonstrate our method's performance against a long list of competing algorithms.
Fichier principal
Vignette du fichier
larsson23a.pdf (2.48 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04211867 , version 1 (20-09-2023)

Licence

Paternité

Identifiants

  • HAL Id : hal-04211867 , version 1

Citer

Johan Larsson, Quentin Klopfenstein, Mathurin Massias, Jonas Wallin. Coordinate Descent for SLOPE. 26th International Conference on Artificial Intelligence and Statistics - AISTATS 2023, Apr 2023, Valencia, Spain. ⟨hal-04211867⟩
24 Consultations
7 Téléchargements

Partager

Gmail Facebook X LinkedIn More