We introduce a method combining variational autoencoders (VAEs) and deep metric learning to perform Bayesian optimisation (BO) over high-dimensional and structured input spaces. By adapting ideas from deep metric learning, we use label guidance from the blackbox function to structure the VAE latent space, facilitating the Gaussian process fit and yielding improved BO performance. Importantly for BO problem settings, our method operates in semi-supervised regimes where only few labelled data points are available. We run experiments on three real-world tasks, achieving state-of-the-art results on the penalised logP molecule generation benchmark using just 3% of the labelled data required by previous approaches. As a theoretical contribution, we present a proof of vanishing regret for VAE BO.
@article{arxiv.2106.03609,
title = {High-Dimensional Bayesian Optimisation with Variational Autoencoders and Deep Metric Learning},
author = {Antoine Grosnit and Rasul Tutunov and Alexandre Max Maraval and Ryan-Rhys Griffiths and Alexander I. Cowen-Rivers and Lin Yang and Lin Zhu and Wenlong Lyu and Zhitang Chen and Jun Wang and Jan Peters and Haitham Bou-Ammar},
journal= {arXiv preprint arXiv:2106.03609},
year = {2021}
}