Communication Dans Un Congrès Année : 2021

Iterative regularization for convex regularizers

Régularisation itérative pour les régulariseurs convexes

Résumé

We study iterative/implicit regularization for linear models, when the bias is convex but not necessarily strongly convex. We characterize the stability properties of a primaldual gradient based approach, analyzing its convergence in the presence of worst case deterministic noise. As a main example, we specialize and illustrate the results for the problem of robust sparse recovery. Key to our analysis is a combination of ideas from regularization theory and optimization in the presence of errors. Theoretical results are complemented by experiments showing that state-of-the-art performances can be achieved with considerable computational speed-ups.

Fichier principal
Vignette du fichier
molinari_massias21a-supp.pdf (650.91 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Licence

Dates et versions

hal-03526107 , version 1 (14-01-2022)

Licence

Identifiants

  • HAL Id : hal-03526107 , version 1

Citer

Cesare Molinari, Mathurin Massias, Lorenzo Rosasco, Silvia Villa. Iterative regularization for convex regularizers. AISTATS 2021, Apr 2021, San Diego / Virtual, United States. ⟨hal-03526107⟩

Collections

62 Consultations
131 Téléchargements

Partager

  • More