Optimization by gradient boosting

Abstract : Gradient boosting is a state-of-the-art prediction technique that sequentially produces a model in the form of linear combinations of simple predictors---typically decision trees---by solving an infinite-dimensional convex optimization problem. We provide in the present paper a thorough analysis of two widespread versions of gradient boosting, and introduce a general framework for studying these algorithms from the point of view of functional optimization. We prove their convergence as the number of iterations tends to infinity and highlight the importance of having a strongly convex risk functional to minimize. We also present a reasonable statistical context ensuring consistency properties of the boosting predictors as the sample size grows. In our approach, the optimization procedures are run forever (that is, without resorting to an early stopping strategy), and statistical regularization is basically achieved via an appropriate $L^2$ penalization of the loss and strong convexity arguments.
Type de document :
Pré-publication, Document de travail
2017
Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-01562618
Contributeur : Gérard Biau <>
Soumis le : dimanche 16 juillet 2017 - 11:53:36
Dernière modification le : samedi 23 septembre 2017 - 01:10:38

Fichiers

biau-cadre.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-01562618, version 1
  • ARXIV : 1707.05023

Citation

Gérard Biau, Benoît Cadre. Optimization by gradient boosting. 2017. 〈hal-01562618〉

Partager

Métriques

Consultations de la notice

304

Téléchargements de fichiers

68