Skip to content
No description, website, or topics provided.
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Type Name Latest commit message Commit time
Failed to load latest commit information.


This repository contains the code for the paper "Screening Rules for Lasso with Non-Convex Sparse Regularizers" published at ICML 2019, Long Beach.

The current version of the paper is available at For abstract, bibliography (.bib), you can look at

this repository is still in construction.


download the source code from the git repository

known dependencies

python (>= 3.6) , scipy (1.1.0), numpy (1.15.4).

structure of the repository

  • contains the code for solving lasso, weighted lasso and proximal weighted lasso using coordinatewise descent

  • contains codes for solving non-convex lasso using coordinate wise descent, with screening, and with screening and screening propagation

  • allows to reproduce experiments using toy data. note that the case (n=500, d=5000) may take several hours especially for bcd

  • allows to generate figures based on the saved results from "compare_algorithms".

for reproducing figure 1 (left) in the paper, run as is and then run

You can’t perform that action at this time.