Skip to Main content Skip to Navigation
Preprints, Working Papers, ...

Quadratic error bound of the smoothed gap and the restarted averaged primal-dual hybrid gradient

Abstract : We study the linear convergence of the primal-dual hybrid gradient method. After a review of current analyses, we show that they do not explain properly the behavior of the algorithm, even on the most simple problems. We thus introduce the quadratic error bound of the smoothed gap, a new regularity assumption that holds for a wide class of optimization problems. Equipped with this tool, we manage to prove tighter convergence rates. Then, we show that averaging and restarting the primal-dual hybrid gradient allows us to leverage better the regularity constant. Numerical experiments on linear and quadratic programs, ridge regression and image denoising illustrate the findings of the paper.
Document type :
Preprints, Working Papers, ...
Complete list of metadata

https://hal.archives-ouvertes.fr/hal-03228252
Contributor : Olivier Fercoq <>
Submitted on : Tuesday, May 18, 2021 - 12:06:27 AM
Last modification on : Sunday, June 6, 2021 - 6:11:10 PM

File

new_regularity_and_rapdhg.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-03228252, version 1

Collections

Citation

Olivier Fercoq. Quadratic error bound of the smoothed gap and the restarted averaged primal-dual hybrid gradient. 2021. ⟨hal-03228252⟩

Share

Metrics

Record views

80

Files downloads

110