In this section you will find EURA NOVA's scientific publications and reports.
In this paper we propose a new method to reduce thesize of Breiman’s Random Forests. Given a RandomForest and a target size, our algorithm builds a lin-ear combination of trees which minimizes the trainingerror. Selected trees, as well as weights of the linearcombination are obtained by means of the OrthogonalMatching Pursuit algorithm. We test our method onmany public benchmark datasets both on regressionand binary classification and we compare it to otherpruning techniques. Experiments show that our tech-nique performs significantly better or equally good onmany datasets1. We also discuss the benefit and short-coming of learning weights for the pruned forest whichlead us to propose to use a non-negative constraint onthe OMP weights for better empirical results.
Luc Giffon, Charly Lamothe, Léo Bouscarrat, Paolo Milanesi, Farah Cherfaoui, Sokol Ko ̧ Pruning Random Forest with Orthogonal Matching Trees, in the proceedings of CAP 2020.