Name
Abstract | ||
|---|---|---|
In this paper, we extend the geometric descent method recently proposed by Bubeck, Lee and Singh [1] to tackle nonsmooth and strongly convex composite problems. We prove that our proposed algorithm, dubbed geometric proximal gradient method (GeoPG), converges with a linear rate (1 - 1 root kappa) and thus achieves the optimal rate among first-order methods, where i is the condition number of the problem. Numerical results on linear regression and logistic regression with elastic net regularization show that GeoPG compares favorably with Nesterov's accelerated proximal gradient method, especially when the problem is ill-conditioned. |
Year | Venue | DocType |
|---|---|---|
2017 | ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 30 (NIPS 2017) | Conference |
Volume | ISSN | Citations |
30 | 1049-5258 | 0 |
PageRank | References | Authors |
0.34 | 0 | 3 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Chen, Shixiang | 1 | 0 | 0.34 |
| Shiqian Ma | 2 | 1068 | 63.48 |
| Wei Liu | 3 | 4041 | 204.19 |