Сравнение методов

	Байесовский стекинг ансамблей	Бустинг	Гауссовский процесс
Область	Машинное обучение	Машинное обучение	Машинное обучение
Семейство	Machine learning	Machine learning	Machine learning
Год появления	2018	1990–1997	2006 (book); roots in Kriging, 1951)
Автор метода	Yao, Y.; Vehtari, A.; Simpson, D.; Gelman, A.	Schapire, R. E.; Freund, Y.	Rasmussen, C. E. & Williams, C. K. I.
Тип	Bayesian ensemble combination	Sequential ensemble (iterative reweighting)	Probabilistic non-parametric model
Основополагающий источник	Yao, Y., Vehtari, A., Simpson, D., & Gelman, A. (2018). Using stacking to average Bayesian predictive distributions. Bayesian Analysis, 13(3), 917–1007.	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.	Rasmussen, C. E., & Williams, C. K. I. (2006). Gaussian Processes for Machine Learning. MIT Press. ISBN: 978-0-262-18253-9
Другие названия	Bayesian stacking, Bayesian model stacking, stacking with Bayesian weights, predictive distribution stacking	AdaBoost, gradient boosting, iterative reweighting ensemble, sequential ensemble	GP, Gaussian Process Regression, GPR, Kriging
Связанные	6	6	3
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.	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.	A Gaussian Process (GP) is a non-parametric, fully probabilistic machine learning model that places a prior distribution directly over functions. Rather than predicting a single value, it returns a predictive mean and a calibrated uncertainty estimate at every test point, making it especially valuable for regression on small to medium datasets and for Bayesian optimization tasks.