Enregistré dans:
Détails bibliographiques
Auteurs principaux: Mayala, Moria, Scornet, Erwan, Tillier, Charles, Wintenberger, Olivier
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:https://arxiv.org/abs/2506.08548
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866912422545063936
author Mayala, Moria
Scornet, Erwan
Tillier, Charles
Wintenberger, Olivier
author_facet Mayala, Moria
Scornet, Erwan
Tillier, Charles
Wintenberger, Olivier
contents Many classification tasks involve imbalanced data, in which a class is largely underrepresented. Several techniques consists in creating a rebalanced dataset on which a classifier is trained. In this paper, we study theoretically such a procedure, when the classifier is a Centered Random Forests (CRF). We establish a Central Limit Theorem (CLT) on the infinite CRF with explicit rates and exact constant. We then prove that the CRF trained on the rebalanced dataset exhibits a bias, which can be removed with appropriate techniques. Based on an importance sampling (IS) approach, the resulting debiased estimator, called IS-ICRF, satisfies a CLT centered at the prediction function value. For high imbalance settings, we prove that the IS-ICRF estimator enjoys a variance reduction compared to the ICRF trained on the original data. Therefore, our theoretical analysis highlights the benefits of training random forests on a rebalanced dataset (followed by a debiasing procedure) compared to using the original data. Our theoretical results, especially the variance rates and the variance reduction, appear to be valid for Breiman's random forests in our experiments.
format Preprint
id arxiv_https___arxiv_org_abs_2506_08548
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Asymptotic Normality of Infinite Centered Random Forests -Application to Imbalanced Classification
Mayala, Moria
Scornet, Erwan
Tillier, Charles
Wintenberger, Olivier
Machine Learning
Many classification tasks involve imbalanced data, in which a class is largely underrepresented. Several techniques consists in creating a rebalanced dataset on which a classifier is trained. In this paper, we study theoretically such a procedure, when the classifier is a Centered Random Forests (CRF). We establish a Central Limit Theorem (CLT) on the infinite CRF with explicit rates and exact constant. We then prove that the CRF trained on the rebalanced dataset exhibits a bias, which can be removed with appropriate techniques. Based on an importance sampling (IS) approach, the resulting debiased estimator, called IS-ICRF, satisfies a CLT centered at the prediction function value. For high imbalance settings, we prove that the IS-ICRF estimator enjoys a variance reduction compared to the ICRF trained on the original data. Therefore, our theoretical analysis highlights the benefits of training random forests on a rebalanced dataset (followed by a debiasing procedure) compared to using the original data. Our theoretical results, especially the variance rates and the variance reduction, appear to be valid for Breiman's random forests in our experiments.
title Asymptotic Normality of Infinite Centered Random Forests -Application to Imbalanced Classification
topic Machine Learning
url https://arxiv.org/abs/2506.08548