Сравнение методов

	Робастное байесовское оценивание	Вариационный вывод
Область	Байесовские методы	Байесовские методы
Семейство	Bayesian methods	Bayesian methods
Год появления	1984–1990	1999
Автор метода	James O. Berger	Jordan, Ghahramani, Jaakkola & Saul
Тип	Bayesian sensitivity / robustness framework	Approximate Bayesian inference
Основополагающий источник	Berger, J. O. (1990). Robust Bayesian analysis: sensitivity to the prior. Journal of Statistical Planning and Inference, 25(3), 303–328.	Jordan, M. I., Ghahramani, Z., Jaakkola, T. S., & Saul, L. K. (1999). An introduction to variational methods for graphical models. Machine Learning, 37(2), 183–233.
Другие названия	Bayesian sensitivity analysis, prior robustness, epsilon-contamination Bayesian analysis, robust Bayes	VI, variational Bayes, VB, mean-field variational inference
Связанные	6	4
Сводка	Robust Bayesian inference extends standard Bayesian analysis by replacing a single prior distribution with a class of plausible priors and examining how much the posterior conclusions change across that class. Instead of committing to one prior, the analyst bounds the posterior quantity of interest, revealing whether findings are stable or critically dependent on prior assumptions.	Variational inference (VI) is a family of techniques that turn Bayesian posterior computation into an optimisation problem. Instead of drawing samples from the exact posterior — as Markov chain Monte Carlo does — VI posits a simpler, tractable family of distributions and finds the member of that family closest to the true posterior by maximising the evidence lower bound (ELBO). Introduced in its modern graphical-model form by Jordan, Ghahramani, Jaakkola and Saul (1999) and given a comprehensive statistical treatment by Blei, Kucukelbir and McAuliffe (2017), VI is now the standard scalable inference engine in probabilistic machine learning.