Beyond L1: Faster and Better Sparse Models with skglm
This addresses the need for faster and more flexible sparse modeling in machine learning, though it appears incremental as it builds on existing methods like coordinate descent.
The paper tackles the problem of estimating sparse generalized linear models with convex or non-convex separable penalties, resulting in an algorithm that solves problems with millions of samples and features in seconds and improves state-of-the-art algorithms.
We propose a new fast algorithm to estimate any sparse generalized linear model with convex or non-convex separable penalties. Our algorithm is able to solve problems with millions of samples and features in seconds, by relying on coordinate descent, working sets and Anderson acceleration. It handles previously unaddressed models, and is extensively shown to improve state-of-art algorithms. We provide a flexible, scikit-learn compatible package, which easily handles customized datafits and penalties.