Distributed Coordinate Descent for L1-regularized Logistic Regression

Ilya Trofimov; Alexander Genkin

doi:10.1007/978-3-319-26123-2_24

paper

Distributed Coordinate Descent for L1-regularized Logistic Regression

arXiv:1411.6520 · doi:10.1007/978-3-319-26123-2_24

Abstract

Solving logistic regression with L1-regularization in distributed settings is an important problem. This problem arises when training dataset is very large and cannot fit the memory of a single machine. We present d-GLMNET, a new algorithm solving logistic regression with L1-regularization in the distributed settings. We empirically show that it is superior over distributed online learning via truncated gradient.