Advances in Operator Splitting Algorithms for Nonsmooth Optimization

Nonsmooth optimization is at the heart of many problems in modern image processing. It turns out that many of these nonsmooth objective functionals have a structure that can be exploited to design fast, effective and provably convergent splitting algorithms. The goal of these algorithms is to achieve full splitting where each part in the objective is used separately. The minisymposium will give an overview of recent advances in operator splitting for nonsmooth optimization, both in the convex and nonconvex case. The minisymposium will bring together recognized experts in optimization theory and imaging sciences.

Organizer: Jalal Fadili
CNRS-ENSICAEN-Université de Caen, France
Gabriel Peyré
Université Paris Dauphine, France
Laurent Condat
CNRS-ENSICAEN-Université de Caen, France

Part I of II

9:30-9:55 Smoothing and First Order Methods: A Unified Framework

Marc Teboulle, Tel Aviv University

10:00-10:25 Implementation of a Block Decomposition Algorithm for Solving Large-scale Conic Optimization Problems

Camilo Ortiz, Georgia Institute of Technology, USA

10:30-10:55 Primal-dual Splitting, Recent Improvements and Variants

Thomas Pock, Graz University of Technology, Austria

Part II of II

2:00-2:25 Fast Alternating Direction Methods

Tom Goldstein , Stanford University, USA

2:30-2:55 Forward-backward Splitting and Proximal Alternating Methods for Nonconvex Nonsmooth Tame Problems

Jérôme Bolte , Université Toulouse I, France

3:00-3:25 A Memory Gradient Algorithm for Nonconvex Regularization with Applications to Image Restoration

Emilie Chouzenoux , Université Paris-Est Marne-la-Vallée, France

3:30-3:55 Combinatorial Selection and Least Absolute Shrinkage via the CLASH Algorithm

Volkan Cevher, Ecole Polytechnique Fédérale de Lausanne, Switzerland