PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Consistency of random forests and other averaging classifiers
Gérard Biau, Luc Devroye and Gábor Lugosi
Journal of Machine Learning Research Volume 9, pp. 2015-2033, 2008.

Abstract

In the last years of his life, Leo Breiman promoted random forests for use in classification. He suggested using averaging as a means of obtaining good discrimination rules. The base classifiers used for averaging are simple and randomized, often based on random samples from the data. He left a few questions unanswered regarding the consistency of such rules. In this paper, we give a number of theorems that establish the universal consistency of averaging rules. We also show that some popular classifiers, including one suggested by Breiman, are not universally consistent.

EPrint Type:Article
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Computational, Information-Theoretic Learning with Statistics
Learning/Statistics & Optimisation
Theory & Algorithms
ID Code:4623
Deposited By:Gábor Lugosi
Deposited On:13 March 2009