8 papers
PACE: Prune-And-Compress Ensemble Models
Fabian Akkerman, Julien Ferry, Théo Guyard +1
Ensemble models achieve state-of-the-art performance on prediction tasks, but usually require aggregating a large number of weak learners. This can hinder deployment, interpretabil…
Linear Decision Tree Policies for Integer Linear Programs
Théo Guyard, Cleber Oliveira, Maximilian Schiffer +2
We study optimal decision policies, represented as linear decision trees, for integer linear programs with a fixed feasible set and varying cost vectors. Once synthesized for a giv…
El0ps: An Exact L0-regularized Problems Solver
Théo Guyard, Cédric Herzet, Clément Elvira
This paper presents El0ps, a Python toolbox providing several utilities to handle L0-regularized problems related to applications in machine learning, statistics, and signal proces…
A Generic Branch-and-Bound Algorithm for -Penalized Problems with Supplementary Material
Clément Elvira, Théo Guyard, Cédric Herzet
We present a generic Branch-and-Bound procedure designed to solve L0-penalized optimization problems. Existing approaches primarily focus on quadratic losses and construct relaxati…
A New Branch-and-Bound Pruning Framework for -Regularized Problems
Theo Guyard, Cédric Herzet, Clément Elvira +1
We consider the resolution of learning problems involving -regularization via Branch-and-Bound (BnB) algorithms. These methods explore regions of the feasible space of the…
Safe Peeling for L0-Regularized Least-Squares with supplementary material
Théo Guyard, Gilles Monnoyer, Clément Elvira +1
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 e…
Node-screening tests for L0-penalized least-squares problem with supplementary material
Théo Guyard, Cédric Herzet, Clément Elvira
We present a novel screening methodology to safely discard irrelevant nodes within a generic branch-and-bound (BnB) algorithm solving the l0-penalized least-squares problem. Our co…
Screen & Relax: Accelerating the resolution of Elastic-net by safe identification of the solution support
Théo Guyard, Cédric Herzet, Clément Elvira
In this paper, we propose a procedure to accelerate the resolution of the well-known "Elastic-Net" problem. Our procedure is based on the (partial) identification of the solution s…