手法を比較

	ベイジアン・スタッキング・アンサンブル	バギング（ブートストラップ集約）	ブースティング
分野	機械学習	機械学習	機械学習
系統	Machine learning	Machine learning	Machine learning
提唱年	2018	1996	1990–1997
提唱者	Yao, Y.; Vehtari, A.; Simpson, D.; Gelman, A.	Breiman, L.	Schapire, R. E.; Freund, Y.
種類	Bayesian ensemble combination	Ensemble meta-algorithm (variance reduction via bootstrap aggregation)	Sequential ensemble (iterative reweighting)
原典	Yao, Y., Vehtari, A., Simpson, D., & Gelman, A. (2018). Using stacking to average Bayesian predictive distributions. Bayesian Analysis, 13(3), 917–1007.	Breiman, L. (1996). Bagging Predictors. Machine Learning, 24(2), 123–140.	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.
別名	Bayesian stacking, Bayesian model stacking, stacking with Bayesian weights, predictive distribution stacking	Bootstrap Aggregating, bootstrap aggregation, bagged ensemble, bagged predictor	AdaBoost, gradient boosting, iterative reweighting ensemble, sequential ensemble
関連	6	5	6
概要	Bayesian stacking combines the predictive distributions of several base models by finding non-negative weights that maximise the leave-one-out log predictive score of the mixture. Formalised by Yao, Vehtari, Simpson, and Gelman (2018), it yields a single calibrated predictive distribution that is provably at least as good as any single constituent model under cross-validation.	Bagging, short for Bootstrap Aggregating, is an ensemble meta-algorithm introduced by Leo Breiman in 1996 that trains multiple copies of a base learner on independently drawn bootstrap samples of the training data and combines their predictions — by averaging for regression or majority vote for classification — to produce a final predictor with substantially lower variance than any single base learner.	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.