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, we introduce several classes of fast EIG estimators by building on ideas from amortized variational inference. We show theoretically and empirically that these estimators can provide significant gains in speed and accuracy over previous approaches. We further demonstrate the practicality of our approach on a number of end-to-end experiments.

Authors

• Adam Foster
• Martin Jankowiak
• Elias Bingham Uber AI Labs
• Paul Horsfall
• Yee Whye Teh
• Thomas Rainforth
• Noah Goodman

Keywords

No keywords are indexed for this paper.