Distributed Coordinate Descent for Generalized Linear Models with Regularization
This addresses the need for efficient distributed training in fields like text mining and clickstream analysis, offering an incremental improvement over existing methods.
The paper tackles the problem of fitting regularized generalized linear models on large, sparse datasets in distributed environments by proposing a novel algorithm that splits data by features, uses coordinate descent locally, and merges results globally with line search, demonstrating scalability and superiority over state-of-the-art approaches in experiments.
Generalized linear model with $L_1$ and $L_2$ regularization is a widely used technique for solving classification, class probability estimation and regression problems. With the numbers of both features and examples growing rapidly in the fields like text mining and clickstream data analysis parallelization and the use of cluster architectures becomes important. We present a novel algorithm for fitting regularized generalized linear models in the distributed environment. The algorithm splits data between nodes by features, uses coordinate descent on each node and line search to merge results globally. Convergence proof is provided. A modifications of the algorithm addresses slow node problem. For an important particular case of logistic regression we empirically compare our program with several state-of-the art approaches that rely on different algorithmic and data spitting methods. Experiments demonstrate that our approach is scalable and superior when training on large and sparse datasets.