# Beyond L1: Faster and Better Sparse Models with skglm

**Authors:** Quentin Bertrand, Quentin Klopfenstein, Pierre-Antoine Bannier, Gauthier Gidel, Mathurin Massias  
**arXiv:** [2204.07826](https://arxiv.org/abs/2204.07826)  
**Date:** 2022-04-16  
**Summary page:** https://www.summarizepaper.com/en/arxiv-id/2204.07826v2/

## Key points

- Authors introduce a novel fast algorithm for estimating sparse generalized linear models with convex or non-convex separable penalties
- Algorithm efficiently solves problems involving millions of samples and features within seconds
- Leveraging techniques such as coordinate descent, working sets, and Anderson acceleration
- Significant improvements over existing state-of-the-art algorithms
- Flexible package compatible with scikit-learn for easy customization of datafits and penalties
- Streamlines implementation process and enhances overall performance of sparse model estimation
- Offers a faster and more efficient approach to handling complex sparse models

## Summary

In their paper titled "Beyond L1: Faster and Better Sparse Models with skglm," authors Quentin Bertrand, Quentin Klopfenstein, Pierre-Antoine Bannier, Gauthier Gidel, and Mathurin Massias introduce a novel fast algorithm for estimating sparse generalized linear models with convex or non-convex separable penalties. This innovative algorithm is capable of efficiently solving problems involving millions of samples and features within seconds. By leveraging techniques such as coordinate descent, working sets, and Anderson acceleration, the proposed algorithm addresses previously unexplored model complexities and demonstrates significant improvements over existing state-of-the-art algorithms. The authors provide a flexible package that is compatible with scikit-learn, enabling easy customization for datafits and penalties. This package not only streamlines the implementation process but also enhances the overall performance of sparse model estimation. The research presented in this paper opens up new possibilities for researchers and practitioners working in the field of statistical machine learning by offering a faster and more efficient approach to handling complex sparse models.