Distributed Coordinate Descent for L1-regularized Logistic Regression
This addresses a practical issue for large-scale machine learning applications, but appears incremental as it builds on existing distributed optimization methods.
The paper tackles the problem of solving L1-regularized logistic regression in distributed settings where datasets are too large for single-machine memory, and presents d-GLMNET, which empirically outperforms distributed online learning via truncated gradient.
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.