Journal Articles SIAM Journal on Optimization Year : 2021

Accelerated Iterative Regularization via Dual Diagonal Descent

Abstract

We propose and analyze an accelerated iterative dual diagonal descent algorithm for the solution of linear inverse problems with strongly convex regularization and general data-fit functions. We develop an inertial approach of which we analyze both convergence and stability properties. Using tools from inexact proximal calculus, we prove early stopping results with optimal convergence rates for additive data terms and further consider more general cases, such as the Kullback--Leibler divergence, for which different type of proximal point approximations hold.

Fichier principal
Vignette du fichier
Fast_3D (2).pdf (800.38 Ko) Télécharger le fichier
Origin Files produced by the author(s)
Licence

Dates and versions

hal-03195616 , version 1 (29-11-2021)

Licence

Identifiers

Cite

Luca Calatroni, Guillaume Garrigos, Lorenzo Rosasco, Silvia Villa. Accelerated Iterative Regularization via Dual Diagonal Descent. SIAM Journal on Optimization, 2021, 31 (1), pp.754-784. ⟨10.1137/19M1308888⟩. ⟨hal-03195616⟩
306 View
220 Download

Altmetric

Share

  • More