We present a generic framework for parallel coordinate descent (CD) algorithms that has as special cases the original sequential algorithms of Cyclic CD and Stochastic CD, as well as the recent parallel Shotgun algorithm of Bradley et al. We introduce two novel parallel algorithms that are also special cases---Thread-Greedy CD and Coloring-Based CD---and give performance measurements for an OpenMP implementation of these.
Revised: October 15, 2012 |
Published: July 3, 2012
Citation
Scherrer C., M. Halappanavar, A. Tewari, and D.J. Haglin. 2012.Scaling Up Coordinate Descent Algorithms for Large l1 Regularization Problems. In Proceedings of the 29th International Conference on Machine Learning (ICML 2012), June 26, 2012, Edinburgh, Scotland, edited by J Langford adn J Pineau. Madison, Wisconsin:International Machine Learning Society.PNNL-SA-87037.