Comparer des méthodes

	Gradient Boosting Robuste	Boosting	Gradient Boosting Régularisé
Domaine	Apprentissage automatique	Apprentissage automatique	Apprentissage automatique
Famille	Machine learning	Machine learning	Machine learning
Année d'origine	2001	1990–1997	2001 (gradient boosting); 2016 (explicit L1/L2 regularization in XGBoost)
Auteur d'origine	Friedman, J. H. (with Huber loss from Huber, P. J.)	Schapire, R. E.; Freund, Y.	Chen, T. & Guestrin, C. (building on Friedman, J. H.)
Type	Ensemble (boosted trees with robust loss)	Sequential ensemble (iterative reweighting)	Regularized ensemble (additive tree model)
Source fondatrice	Friedman, J. H. (2001). Greedy function approximation: A gradient boosting machine. Annals of Statistics, 29(5), 1189–1232.	Freund, Y. & Schapire, R. E. (1997). A decision-theoretic generalization of on-line learning and an application to boosting. Journal of Computer and System Sciences, 55(1), 119–139.	Chen, T. & Guestrin, C. (2016). XGBoost: A scalable tree boosting system. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 785–794.
Alias	gradient boosting with Huber loss, robust GBM, outlier-robust boosting, robust gradient-boosted trees	AdaBoost, gradient boosting, iterative reweighting ensemble, sequential ensemble	penalized gradient boosting, shrinkage-regularized boosting, XGBoost-style regularization, L1/L2 gradient boosting
Apparentées	6	6	6
Résumé	Robust Gradient Boosting is gradient boosting trained with outlier-resistant loss functions — most commonly the Huber loss or quantile (pinball) loss — instead of squared-error loss. Proposed in Friedman's seminal 2001 paper, this variant produces predictions far less distorted by extreme values or contaminated labels, while retaining the full predictive power of gradient-boosted trees.	Boosting is a sequential ensemble technique that converts many simple, barely-better-than-chance learners into a single highly accurate model by repeatedly focusing training on the examples that previous learners got wrong, then combining all learners with weights proportional to their individual accuracy.	Regularized gradient boosting extends the classic additive tree ensemble (Friedman 2001) by embedding L1 and L2 penalty terms directly into the training objective, along with a complexity penalty on tree size. Popularized by XGBoost (Chen & Guestrin 2016), this framework reduces overfitting and improves generalization compared to unpenalized boosting, while retaining the method's characteristic accuracy on tabular data.