Implicit Deep Adaptive Design: Policy-Based Experimental Design without Likelihoods

Desi R. Ivanova, Adam Foster, Steven Kleinegesse, Michael U. Gutmann, Tom Rainforth

Keywords: Experimental Design, Mutual Information, Implicit Models, Bayesian methods, Bayesian Optimal Experimental Design, BOED, Sequential BOED, Adaptive experiments, Likelihood-free methods, Likelihood-free inference, Parameter estimation, Variational methods, Mutual information lower bounds, Stochastic simulator models

TL;DR: A new policy-based method for performing Bayesian experimental design with implicit models that does not require heavy computations during the experiment, opening the door to running adaptive experiments in real time.

Abstract: We introduce implicit Deep Adaptive Design (iDAD), a new method for performing adaptive experiments in real-time with implicit models. iDAD amortizes the cost of Bayesian optimal experimental design (BOED) by learning a design policy network upfront, which can then be deployed quickly at the time of the experiment. The iDAD network can be trained on any model which simulates differentiable samples, unlike previous design policy work that requires a closed form likelihood and conditionally independent experiments. At deployment, iDAD allows design decisions to be made in milliseconds, in contrast to traditional BOED approaches that require heavy computation during the experiment itself. We illustrate the applicability of iDAD on a number of experiments, and show that it provides a fast and effective mechanism for performing adaptive design with implicit models.

Code Of Conduct: I certify that all co-authors of this work have read and commit to adhering to the NeurIPS Statement on Ethics, Fairness, Inclusivity, and Code of Conduct.

Code: https://github.com/desi-ivanova/idad