Variance reduction in purely random forests - Archive ouverte HAL Access content directly
Journal Articles Journal of Nonparametric Statistics Year : 2012

Variance reduction in purely random forests

(1)
1

Abstract

Random forests, introduced by Leo Breiman in 2001, are a very effective statistical method. The complex mechanism of the method makes theoretical analysis difficult. Therefore, simplified versions of random forests, called purely random forests, which can be theoretically handled more easily, have been considered. In this paper we study the variance of such forests. First, we show a general upper bound which emphasizes the fact that a forest reduces the variance. We then introduce a simple variant of purely random forests, that we call purely uniformly random forests. For this variant and in the context of regression problems with a one-dimensional predictor space, we show that both random trees and random forests reach minimax rate of convergence. In addition, we prove that compared to random trees, random forests improve accuracy by reducing the estimator variance by a factor of three fourths.
Fichier principal
Vignette du fichier
genuer.var-reduc-prf-preprint.pdf (221.85 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-01590513 , version 1 (19-09-2017)

Licence

Attribution - CC BY 4.0

Identifiers

Cite

Robin Genuer. Variance reduction in purely random forests. Journal of Nonparametric Statistics, 2012, 2, pp.18 - 562. ⟨10.1007/978-1-4899-0027-2⟩. ⟨hal-01590513⟩

Collections

INSMI
119 View
1605 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More