L0Learn: A Scalable Package for Sparse Learning using L0 Regularization

2022/02/10 by Hussein Hazimeh, Rahul Mazumder, Hazimeh, Hussein +3 · 5 citations
Computer Science · #Advanced Data Compression Techniques #Computation (stat.CO) #FOS: Computer and information sciences #Gaussian Processes and Bayesian Inference #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Machine Learning and Data Classification #Mathematical Software (cs.MS)

paper · pdf · doi:10.48550/arxiv.2202.04820

openalex publication_date 2022/02/10 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

We present L0Learn: an open-source package for sparse linear regression and classification using ℓ₀ regularization. L0Learn implements scalable, approximate algorithms, based on coordinate descent and local combinatorial optimization. The package is built using C++ and has user-friendly R and Python interfaces. L0Learn can address problems with millions of features, achieving competitive run times and statistical performance with state-of-the-art sparse learning packages. L0Learn is available on both CRAN and GitHub (https://cran.r-project.org/package=L0Learn and https://github.com/hazimehh/L0Learn).

Cited by

• A Generic Branch-and-Bound Algorithm for ℓ₀-Penalized Problems with Supplementary Material
• El0ps: An Exact L0-regularized Problems Solver
• MOSS: Multi-Objective Optimization for Stable Rule Sets
• Foundations of Top-k Decoding For Language Models
• Nash: Neural Adaptive Shrinkage for Structured High-Dimensional Regression

Related

• El0ps: An Exact L0-regularized Problems Solver
• Learning Sparse Classifiers: Continuous and Mixed Integer Optimization Perspectives
• Fast Best Subset Selection: Coordinate Descent and Local Combinatorial\n Optimization Algorithms
• Lass-0: sparse non-convex regression by local search
• Sparsifying Parametric Models with L0 Regularization