Parallel block coordinate descent methods with identification strategies
Abstract
This work presents a parallel variant of the algorithm introduced in [Acceleration of block coordinate descent methods with identification strategies Comput. Optim. Appl. 72(3):609--640, 2019] to minimize the sum of a partially separable smooth convex function and a possibly non-smooth block-separable convex function under simple constraints. It achieves better efficiency by using a strategy to identify the nonzero coordinates that allows the computational effort to be focused on using a nonuniform probability distribution in the selection of the blocks. Parallelization is achieved by extending the theoretical results from Richt\'arik and Tak\'a\v{c} [Parallel coordinate descent methods for big data optimization, Math. Prog. Ser. A 156:433--484, 2016]. We present convergence results and comparative numerical experiments on regularized regression problems using both synthetic and real data.
Cite
@article{arxiv.2507.22277,
title = {Parallel block coordinate descent methods with identification strategies},
author = {Ronaldo Lopes and Sandra A. Santos and Paulo J. S. Silva},
journal= {arXiv preprint arXiv:2507.22277},
year = {2025}
}
Comments
56 pages (with an appendix with all running times in tables), 12 figures