Comparer des méthodes

	Boosting Régularisé	Boosting
Domaine	Apprentissage automatique	Apprentissage automatique
Famille	Machine learning	Machine learning
Année d'origine	2001–2016	1990–1997
Auteur d'origine	Friedman, J. H.; extended by Chen & Guestrin	Schapire, R. E.; Freund, Y.
Type	Regularized ensemble (boosting with shrinkage/penalty)	Sequential ensemble (iterative reweighting)
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.
Alias	shrinkage boosting, penalized boosting, regularized gradient boosting, L1/L2 boosting	AdaBoost, gradient boosting, iterative reweighting ensemble, sequential ensemble
Apparentées	5	6
Résumé	Regularized boosting extends gradient boosting by adding explicit controls — shrinkage (learning rate), L1/L2 weight penalties, subsampling, and tree-complexity limits — to the objective function and the update rule. These constraints reduce overfitting, stabilise the model on noisy or small datasets, and are the core reason why systems such as XGBoost and LightGBM consistently outperform vanilla boosting on real-world tabular benchmarks.	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.