Ranking over Regression for Bayesian Optimization and Molecule Selection

TL;DR

This paper introduces Rank-based Bayesian Optimization (RBO), a novel approach that uses ranking models as surrogates, showing improved performance over traditional regression-based BO in chemical compound optimization, especially with complex property landscapes.

Contribution

The paper proposes RBO, a new BO method utilizing ranking models as surrogates, and demonstrates its effectiveness for chemical molecule optimization over traditional regression models.

Findings

RBO achieves similar or better optimization results compared to conventional BO.

Ranking surrogate models perform well even at early iterations of BO.

High correlation exists between ranking ability and BO performance.

Abstract

Bayesian optimization (BO) has become an indispensable tool for autonomous decision-making across diverse applications from autonomous vehicle control to accelerated drug and materials discovery. With the growing interest in self-driving laboratories, BO of chemical systems is crucial for machine learning (ML) guided experimental planning. Typically, BO employs a regression surrogate model to predict the distribution of unseen parts of the search space. However, for the selection of molecules, picking the top candidates with respect to a distribution, the relative ordering of their properties may be more important than their exact values. In this paper, we introduce Rank-based Bayesian Optimization (RBO), which utilizes a ranking model as the surrogate. We present a comprehensive investigation of RBO's optimization performance compared to conventional BO on various chemical datasets. Our results demonstrate similar or improved optimization performance using ranking models, particularly for datasets with rough structure-property landscapes and activity cliffs. Furthermore, we observe a high correlation between the surrogate ranking ability and BO performance, and this ability is maintained even at early iterations of BO optimization when using ranking surrogate models. We conclude that RBO is an effective alternative to regression-based BO, especially for optimizing novel chemical compounds.