Variational Bayesian optimal experimental design
Bayesian optimal experimental design (BOED) is a principled framework for making efficient use of limited experimental resources. Unfortunately, its applicability is hampered by the difficulty of obtaining accurate estimates of the expected information gain (EIG) of an experiment. To address this, w...
Main Authors: | Foster, A, Jankowiak, M, Bingham, E, Horsfall, P, Tee, YW, Rainforth, T, Goodman, N |
---|---|
Formato: | Conference item |
Publicado em: |
Conference on Neural Information Processing Systems
2019
|
Registos relacionados
-
A unified stochastic gradient approach to designing Bayesian-optimal experiments
Por: Foster, A, et al.
Publicado em: (2020) -
Modern Bayesian experimental design
Por: Rainforth, T, et al.
Publicado em: (2024) -
Variational, Monte Carlo and policy-based approaches to Bayesian experimental design
Por: Foster, AE
Publicado em: (2021) -
Making better use of unlabelled data in Bayesian Active learning
Por: Bickford Smith, F, et al.
Publicado em: (2024) -
Bayesian Optimization for Probabilistic Programs
Por: Rainforth, T, et al.
Publicado em: (2016)