Сравнение методов

	Байесовский стекинг ансамблей	Байесовское усреднение моделей	Бустинг
Область	Машинное обучение	Байесовские методы	Машинное обучение
Семейство	Machine learning	Bayesian methods	Machine learning
Год появления	2018	1999	1990–1997
Автор метода	Yao, Y.; Vehtari, A.; Simpson, D.; Gelman, A.	Hoeting, Madigan, Raftery & Volinsky	Schapire, R. E.; Freund, Y.
Тип	Bayesian ensemble combination	Bayesian model averaging	Sequential ensemble (iterative reweighting)
Основополагающий источник	Yao, Y., Vehtari, A., Simpson, D., & Gelman, A. (2018). Using stacking to average Bayesian predictive distributions. Bayesian Analysis, 13(3), 917–1007.	Hoeting, J. A., Madigan, D., Raftery, A. E. & Volinsky, C. T. (1999). Bayesian Model Averaging: A Tutorial. Statistical Science, 14(4), 382–401.	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.
Другие названия	Bayesian stacking, Bayesian model stacking, stacking with Bayesian weights, predictive distribution stacking	BMA, Bayesian model combination, Bayesian Model Ortalaması (BMA)	AdaBoost, gradient boosting, iterative reweighting ensemble, sequential ensemble
Связанные	6	5	6
Сводка	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.	Bayesian Model Averaging (BMA), formalised as a tutorial by Hoeting, Madigan, Raftery and Volinsky in 1999, addresses model uncertainty by averaging over all plausible model specifications rather than selecting a single best model. Each candidate model receives a posterior probability that reflects how well it fits the data given a prior, and predictions or coefficient estimates are formed as weighted averages across the entire model space. This approach reduces the bias and overconfidence that arise when a single selected model is treated as the true one.	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.