High-Dimensional Bayesian Optimisation with Variational Autoencoders and Deep Metric Learning
arXiv:2106.03609
Abstract
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.
References in corpus (10)
- A Tutorial on Bayesian Optimization of Expensive Cost Functions, with Application to Active User Modeling and Hierarchical Reinforcement Learning
- CLEAR: Contrastive Learning for Sentence Representation
- Scalable Global Optimization via Local Bayesian Optimization
- BayesNAS: A Bayesian Approach for Neural Architecture Search
- All SMILES Variational Autoencoder
- Improving black-box optimization in VAE latent space using decoder uncertainty
- JANUS: Parallel Tempered Genetic Algorithm Guided by Deep Neural Networks for Inverse Molecular Design
- Are we Forgetting about Compositional Optimisers in Bayesian Optimisation?
- Adaptive Sensor Placement for Continuous Spaces
- Good practices for Bayesian Optimization of high dimensional structured spaces