%0 Journal Article
%T Fast Distributed Coordinate Descent for Non-Strongly Convex Losses
%A Olivier Fercoq
%A Zheng Qu
%A Peter Richt¨¢rik
%A Martin Tak¨¢£¿
%J Mathematics
%D 2014
%I arXiv
%X We propose an efficient distributed randomized coordinate descent method for minimizing regularized non-strongly convex loss functions. The method attains the optimal $O(1/k^2)$ convergence rate, where $k$ is the iteration counter. The core of the work is the theoretical study of stepsize parameters. We have implemented the method on Archer - the largest supercomputer in the UK - and show that the method is capable of solving a (synthetic) LASSO optimization problem with 50 billion variables.
%U http://arxiv.org/abs/1405.5300v2