Xudong Li Princeton University, Princeton

Defeng Sun The Hong Kong Polytechnic University, Hung Hom, Hong Kong

Kim Chuan Toh Department of Mathematics and Institute of Operations Research and Analytics, National University of Singapore

ABSTRACT

We develop a fast and robust algorithm for solving large-scale convex composite optimization models with an emphasis on the 1-regularized least squares regression (lasso) problems. Despite the fact that there exists a large number of solvers in the literature for the lasso problems, we found that no solver can efficiently handle difficult large-scale regression problems with real data.
By leveraging on available error bound results to realize the asymptotic super linear convergence property of the augmented Lagrangian algorithm, and by exploiting the second order sparsity of the problem through the semismooth Newton method, we are able to propose an algorithm, called SSNAL, to efficiently solve the aforementioned difficult problems. Under very mild conditions, which hold automatically for lasso problems, both the primal and the dual iteration sequences generated by SSNAL possess a fast linear convergence rate, which can even be super linear asymptotically. Numerical comparisons between our approach and a number of state-of-the-art solvers, on real data sets, are presented to demonstrate the high efficiency and robustness of our proposed algorithm in solving difficult large-scale lasso problems.