Best Practices for Multi-Fidelity Bayesian Optimization in Materials and Molecular Research

Victor Sabanza Gil, Riccardo Barbano, Daniel Pacheco Gutiérrez, Jeremy Scott Luterbacher, José Miguel Hernández-Lobato, Philippe Schwaller, Loïc Roch

Keywords: molecules, bayesian optimization, multi-fidelity

TL;DR: Guidelines for using Multi-fidelity Bayesian Optimization (MFBO) in experimental settings, benchmarking its effectiveness in molecular and materials discovery tasks and comparing it to single-fidelity approaches.

Abstract:

Multi-fidelity Bayesian Optimization (MFBO) is a promising framework to speed
up materials and molecular discovery as sources of information of different accuracies are at hand at increasing cost. Despite its potential use in chemical tasks,
there is a lack of systematic evaluation of the many parameters playing a role
in MFBO. In this work, we provide guidelines and recommendations to decide
when to use MFBO in experimental settings. We investigate MFBO methods
applied to molecules and materials problems. First, we test two different families
of acquisition functions in two synthetic problems and study the effect of the
informativeness and cost of the approximate function. We use our implementation
and guidelines to benchmark three real discovery problems and compare them
against their single-fidelity counterparts. Our results may help guide future efforts
to implement MFBO as a routine tool in the chemical sciences.

Submission Number: 58