Safe Peeling for L0-Regularized Least-Squares with supplementary material
Machine Learning
2023-06-07 v4 Optimization and Control
Machine Learning
Abstract
We introduce a new methodology dubbed ``safe peeling'' to accelerate the resolution of L0-regularized least-squares problems via a Branch-and-Bound (BnB) algorithm. Our procedure enables to tighten the convex relaxation considered at each node of the BnB decision tree and therefore potentially allows for more aggressive pruning. Numerical simulations show that our proposed methodology leads to significant gains in terms of number of nodes explored and overall solving time.s show that our proposed methodology leads to significant gains in terms of number of nodes explored and overall solving time.
Cite
@article{arxiv.2302.14471,
title = {Safe Peeling for L0-Regularized Least-Squares with supplementary material},
author = {Théo Guyard and Gilles Monnoyer and Clément Elvira and Cédric Herzet},
journal= {arXiv preprint arXiv:2302.14471},
year = {2023}
}