Сравнение методов

	Динамическое байесовское усреднение моделей	Динамический вариационный вывод
Область	Байесовские методы	Байесовские методы
Семейство	Bayesian methods	Bayesian methods
Год появления	2010	2014–2015
Автор метода	Raftery, Karny & Ettler	Bayer, Osendorfer, Krishnan and colleagues
Тип	dynamic ensemble / model combination	Bayesian approximate inference
Основополагающий источник	Raftery, A. E., Karny, M., & Ettler, P. (2010). Online prediction under model uncertainty via dynamic model averaging: Application to a cold rolling mill. Technometrics, 52(1), 52-66.	Krishnan, R. G., Shalit, U., & Sontag, D. (2015). Deep Kalman Filters. NIPS 2015 Workshop on Advances in Approximate Bayesian Inference.
Другие названия	DMA, dynamic model averaging, time-varying BMA, online Bayesian model averaging	sequential variational inference, temporal variational inference, variational inference for state-space models, DVI
Связанные	6	6
Сводка	Dynamic Bayesian Model Averaging (DMA) extends standard Bayesian model averaging to settings where the best predictive model may change over time. It maintains a probability distribution over a set of competing models and updates that distribution sequentially as new observations arrive, allowing model weights to evolve rather than remaining fixed across the entire sample.	Dynamic variational inference extends the variational inference framework to sequential and time-series settings by positing a structured approximate posterior that respects the temporal ordering of latent states. It jointly learns a generative model of how hidden states evolve over time and a recognition network that maps observed sequences back to those latent states, optimising a sequential evidence lower bound (ELBO).