A Framework for Bayesian Optimization in Mixture Spaces

Paola Driza, Benjamin Sanchez-Lengeling

Submission Type: I want my submission to be considered for both oral and poster presentation.

Keywords: Mixtures, bayesian optimization, self-driving labs, AI for materials science, constrained optimization

TL;DR: We propose a constraint taxonomy for informing constraint-based BO algorithms in the mixture formulation space, by formalizing mixture design as compositional accords under practical and chemical constraints.

Confirmation Of Submission Requirements: I submit an abstract. It uses the template provided on the submission page and is no longer than 2 pages.

Submission Number: 416