Faithful inversion of generative models for effective amortized inference
Inference amortization methods share information across multiple posteriorinference problems, allowing each to be carried out more efficiently. Generally, they require the inversion of the dependency structure in the generative model, as the modeller must learn a mapping from observations to distrib...
المؤلفون الرئيسيون: | Webb, S, Golinski, A, Zinkov, R, Narayanaswamy, S, Rainforth, T, Teh, Y, Wood, F |
---|---|
التنسيق: | Conference item |
منشور في: |
Curran Associates
2019
|
مواد مشابهة
-
Amortized Monte Carlo integration
حسب: Goliński, A, وآخرون
منشور في: (2019) -
Amortized rejection sampling in universal probabilistic programming
حسب: Naderiparizi, S, وآخرون
منشور في: (2022) -
Amortized inference and model learning for probabilistic programming
حسب: Le, TA
منشور في: (2019) -
Kilonova Spectral Inverse Modelling with Simulation-based Inference: An Amortized Neural Posterior Estimation Analysis
حسب: P. Darc, وآخرون
منشور في: (2024-01-01) -
Target–aware Bayesian inference: how to beat optimal conventional estimators
حسب: Rainforth, T, وآخرون
منشور في: (2020)