方法对比

	贝叶斯装袋法	贝叶斯随机森林	Boosting
领域	机器学习	机器学习	机器学习
方法族	Machine learning	Machine learning	Machine learning
起源年份	2001	2015	1990–1997
提出者	Clyde, M. & Lee, H. (building on Rubin's Bayesian bootstrap, 1981)	Taddy, M. et al.	Schapire, R. E.; Freund, Y.
类型	Ensemble (Bayesian bootstrap aggregation)	Bayesian ensemble of decision trees	Sequential ensemble (iterative reweighting)
开创性文献	Clyde, M. & Lee, H. (2001). Bagging and the Bayesian bootstrap. In T. Richardson & T. Jaakkola (Eds.), Proceedings of the Eighth International Workshop on Artificial Intelligence and Statistics (AISTATS 2001).	Taddy, M., Chen, C., Yu, J., & Wyle, M. (2015). Bayesian and Empirical Bayesian Forests. Proceedings of the 32nd International Conference on Machine Learning (ICML 2015), PMLR 37, 967–976.	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.
别名	Bayesian bootstrap aggregation, BB-ensemble, Bayesian model averaging via bootstrap, Bayesian bagged ensemble	Bayesian Forest, BRF, Empirical Bayesian Forest, posterior random forest	AdaBoost, gradient boosting, iterative reweighting ensemble, sequential ensemble
相关	6	5	6
摘要	Bayesian Bagging replaces the classical bootstrap with the Bayesian bootstrap — drawing Dirichlet-distributed weights over training observations rather than sampling with replacement — and trains an ensemble of base learners under those weights. The result is a principled ensemble that approximates a Bayesian posterior over predictions, yielding calibrated uncertainty estimates alongside strong predictive accuracy.	Bayesian Random Forest extends the classical random forest by placing a prior distribution over tree structures and leaf parameters, then sampling or approximating the posterior over that ensemble. The result is a set of predictions accompanied by calibrated uncertainty estimates — a capability standard random forests lack — making it valuable when knowing how confident the model is matters as much as the prediction itself.	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.