Oral
in
Workshop: Bayesian Decision-making and Uncertainty: from probabilistic and spatiotemporal modeling to sequential experiment design
Variational Search Distributions
Daniel Steinberg · Rafael Oliveira · Cheng Soon Ong · Edwin Bonilla
Keywords: [ Variational Inference ] [ Bayesian optimization ] [ Active Learning ]
We develop variational search distributions (VSD), a method for finding designs of a rare desired class in a batch sequential manner with a set experimental budget. We formalise the requirements and desiderata for this problem and formulate a solution via variational inference that fulfil these. In particular, VSD uses off-the-shelf gradient based optimisation routines, and can take advantage of scalable predictive models. We show that VSD can outperform existing baseline methods on a set of real sequence-design problems in various biological systems.