Efficiency of randomized coordinate descent methods on minimization problems with a composite objective function
暂无分享,去创建一个
We develop a randomized block-coordinate descent method for minimizing the sum of a smooth and a simple nonsmooth blockseparable convex function and prove that it obtains an ε-accurate solution with probability at least 1−ρ in at most O((4n/ε) log(1/ερ)) iterations, where n is the dimension of the problem. This extends recent results of Nesterov [2], which cover the smooth case, to composite minimization, and improves the complexity by a factor of 2. In the smooth case we give a much simplified analysis. Finally, we demonstrate numerically that the algorithm is able to solve various l1-regularized optimization problems with a billion variables.
[1] Yurii Nesterov,et al. Efficiency of Coordinate Descent Methods on Huge-Scale Optimization Problems , 2012, SIAM J. Optim..
[2] Chih-Jen Lin,et al. Coordinate Descent Method for Large-scale L2-loss Linear Support Vector Machines , 2008, J. Mach. Learn. Res..
[3] Stephen J. Wright,et al. Sparse Reconstruction by Separable Approximation , 2008, IEEE Transactions on Signal Processing.