Shotgun: Parallel Coordinate Descent for L1-Regularized Loss Minimization

Authors

Please contact any of the authors for further information.

Abstract

We propose Shotgun, a parallel coordinate descent algorithm for minimizing L1-regularized losses. Though coordinate descent seems inherently sequential, we prove convergence bounds for Shotgun which predict linear speedups, up to a problem-dependent limit. We present a comprehensive empirical study of Shotgun for Lasso and sparse logistic regression. Our theoretical predictions on the potential for parallelism closely match behavior on real data. Shotgun outperforms other published solvers on a range of large problems, proving to be one of the most scalable algorithms for L1.

Code

Please see the Shotgun section on the Select Lab's Code page.