Bayesian Optimisation: Which Constraints Matter?

TL;DR

This paper introduces new Bayesian optimisation methods that efficiently handle decoupled black-box constraints by focusing evaluations on relevant constraints, improving optimization performance.

Contribution

It proposes Bayesian optimisation variants of Knowledge Gradient for problems with decoupled constraints, emphasizing evaluation of only binding constraints.

Findings

Methods outperform existing approaches in benchmarks.

Focusing on relevant constraints improves optimization efficiency.

Demonstrates superiority over state-of-the-art techniques.

Abstract

Bayesian optimisation has proven to be a powerful tool for expensive global black-box optimisation problems. In this paper, we propose new Bayesian optimisation variants of the popular Knowledge Gradient acquisition functions for problems with \emph{decoupled} black-box constraints, in which subsets of the objective and constraint functions may be evaluated independently. In particular, our methods aim to take into account that often only a handful of the constraints may be binding at the optimum, and hence we should evaluate only relevant constraints when trying to optimise a function. We empirically benchmark these methods against existing methods and demonstrate their superiority over the state-of-the-art.