So sánh phương pháp

	Tăng cường Bayes	Boosting	Semi-supervised Boosting
Lĩnh vực	Học máy	Học máy	Học máy
Họ	Machine learning	Machine learning	Machine learning
Năm ra đời	1999–2010	1990–1997	1999–2009
Người khởi xướng	Ridgeway, G.; Chipman, H. A. et al.	Schapire, R. E.; Freund, Y.	Mallapragada, P. K.; Bennett, K. P.; and others
Loại	Probabilistic ensemble (Bayesian interpretation of boosting)	Sequential ensemble (iterative reweighting)	Semi-supervised ensemble method
Công trình gốc	Ridgeway, G. (1999). The state of boosting. Computing Science and Statistics, 31, 172–181.	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.	Mallapragada, P. K., Jin, R., Jain, A. K., & Liu, Y. (2009). SemiBoost: Boosting for Semi-supervised Learning. IEEE Transactions on Pattern Analysis and Machine Intelligence, 31(11), 2000–2014.
Tên gọi khác	Bayesian ensemble boosting, probabilistic boosting, Bayesian additive model, Bayesian boosted ensemble	AdaBoost, gradient boosting, iterative reweighting ensemble, sequential ensemble	SemiBoost, SSL boosting, boosting with unlabeled data, semi-supervised ensemble boosting
Liên quan	5	6	5
Tóm tắt	Bayesian boosting integrates probabilistic Bayesian inference with boosting ensemble techniques, combining multiple weak learners while maintaining full uncertainty quantification over predictions. Unlike standard gradient boosting that produces a single point estimate, Bayesian boosting yields a posterior distribution over the ensemble output, enabling calibrated confidence intervals alongside predictions.	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.	Semi-supervised Boosting is an ensemble learning paradigm that extends classical boosting algorithms — such as AdaBoost — to exploit both labeled and unlabeled data. By propagating label information through a similarity structure over unlabeled instances, it trains stronger classifiers than supervised boosting alone when labeled data are scarce.