On the Linear Convergence of the Approximate Proximal Splitting Method for Non-Smooth Convex Optimization
arXiv:1404.5350
Abstract
Consider the problem of minimizing the sum of two convex functions, one being smooth and the other non-smooth. In this paper, we introduce a general class of approximate proximal splitting (APS) methods for solving such minimization problems. Methods in the APS class include many well-known algorithms such as the proximal splitting method (PSM), the block coordinate descent method (BCD) and the approximate gradient projection methods for smooth convex optimization. We establish the linear convergence of APS methods under a local error bound assumption. Since the latter is known to hold for compressive sensing and sparse group LASSO problems, our analysis implies the linear convergence of the BCD method for these problems without strong convexity assumption.
arXiv comment21 pages, no figures
References in corpus (4)
- Consistency of the group Lasso and multiple kernel learning
- Iteration Complexity Analysis of Block Coordinate Descent Methods
- Incremental Majorization-Minimization Optimization with Application to Large-Scale Machine Learning
- Parallel coordinate descent methods for composite minimization: convergence analysis and error bounds