AUGMENTED LAGRANGIAN WITHOUT ALTERNATING DIRECTIONS: PRACTICAL ALGORITHMS FOR INVERSE PROBLEMS IN IMAGING - Université Jean-Monnet-Saint-Étienne Access content directly
Conference Papers Year : 2015

AUGMENTED LAGRANGIAN WITHOUT ALTERNATING DIRECTIONS: PRACTICAL ALGORITHMS FOR INVERSE PROBLEMS IN IMAGING

Abstract

Several problems in signal processing and machine learning can be casted as optimization problems. In many cases, they are of large-scale, nonlinear, have constraints, and may be nonsmooth in the unknown parameters. There exists plethora of fast algorithms for smooth convex optimization, but these algorithms are not readily applicable to nonsmooth problems , which has led to a considerable amount of research in this direction. In this paper, we propose a general algorithm for nonsmooth bound-constrained convex optimization problems. Our algorithm is instance of the so-called augmented Lagrangian, for which theoretical convergence is well established for convex problems. The proposed algorithm is a blend of superlinearly convergent limited memory quasi-Newton method, and proximal projection operator. The initial promising numerical results for total-variation based image deblurring show that they are as fast as the best existing algorithms in the same class, but with fewer and less sensitive tuning parameters, which makes a huge difference in practice.
Fichier principal
Vignette du fichier
388-Bt3f-151.pdf (302.91 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

ujm-01122878 , version 1 (04-03-2015)

Identifiers

  • HAL Id : ujm-01122878 , version 1

Cite

Rahul Mourya, Loic Denis, Éric Thiébaut, Jean-Marie Becker. AUGMENTED LAGRANGIAN WITHOUT ALTERNATING DIRECTIONS: PRACTICAL ALGORITHMS FOR INVERSE PROBLEMS IN IMAGING. IEEE International Conference on Image Processing (ICIP), Sep 2015, Quebec, Canada. ⟨ujm-01122878⟩
263 View
813 Download

Share

Gmail Facebook Twitter LinkedIn More