Think Global and Act Local: Bayesian Optimisation over High-Dimensional Categorical and Mixed Search Spaces

Abstract

High-dimensional black-box optimisation remains an important yet notoriously challenging problem. Despite the success of Bayesian optimisation methods on continuous domains, domains that are categorical, or that mix continuous and categorical variables, remain challenging. We propose a novel solution – we combine local optimisation with a tailored kernel design, effectively handling high-dimensional categorical and mixed search spaces, whilst retaining sample efficiency. We further derive convergence guarantee for the proposed approach. Finally, we demonstrate empirically that our method outperforms the current baselines on a variety of synthetic and real-world tasks in terms of performance, computational costs, or both.

Publication
Proceedings of the 38th International Conference on Machine Learning
Xingchen Wan
Xingchen Wan
Research Scientist

My research interests include Bayesian optimization, AutoML, graphs & large language models.