Timezone: »

 
Poster
A Multi-step Inertial Forward-Backward Splitting Method for Non-convex Optimization
Jingwei Liang · Jalal Fadili · Gabriel Peyré

Wed Dec 07 09:00 AM -- 12:30 PM (PST) @ Area 5+6+7+8 #73
in Wednesday Posters »

In this paper, we propose a multi-step inertial Forward--Backward splitting algorithm for minimizing the sum of two non-necessarily convex functions, one of which is proper lower semi-continuous while the other is differentiable with a Lipschitz continuous gradient. We first prove global convergence of the scheme with the help of the Kurdyka–Łojasiewicz property. Then, when the non-smooth part is also partly smooth relative to a smooth submanifold, we establish finite identification of the latter and provide sharp local linear convergence analysis. The proposed method is illustrated on a few problems arising from statistics and machine learning.

Keywords: (Other) Optimization · Sparsity and Feature Selection · Nonlinear Dimension Reduction and Manifold Learning
[ PDF] 

Author Information

Jingwei Liang (GREYC)
Jalal Fadili (CNRS-ENSICAEN-Univ. Caen)
Gabriel Peyré (Université Paris Dauphine)

More from the Same Authors