A Second-Order Method for Strongly Convex L1-Regularization Problems
Optimization and Control
2015-01-13 v6
Abstract
In this paper a robust second-order method is developed for the solution of strongly convex l1-regularized problems. The main aim is to make the proposed method as inexpensive as possible, while even difficult problems can be efficiently solved. The proposed approach is a primal-dual Newton Conjugate Gradients (pdNCG) method. Convergence properties of pdNCG are studied and worst-case iteration complexity is established. Numerical results are presented on synthetic sparse least-squares problems and real world machine learning problems.
Cite
@article{arxiv.1306.5386,
title = {A Second-Order Method for Strongly Convex L1-Regularization Problems},
author = {Kimon Fountoulakis and Jacek Gondzio},
journal= {arXiv preprint arXiv:1306.5386},
year = {2015}
}
Comments
30 pages, 13 figures, 1 table