Exploiting gradients and Hessians in Bayesian optimization and Bayesian quadrature

Abstract

An exciting branch of machine learning research focuses on methods for learning, optimizing, and integrating unknown functions that are difficult or costly to evaluate. A popular Bayesian approach to this problem uses a Gaussian process (GP) to construct a posterior distribution over the function of interest given a set of observed measurements, and selects new points to evaluate using the statistics of this posterior. Here we extend these methods to exploit derivative information from the unknown function. We describe methods for Bayesian optimization (BO) and Bayesian quadrature (BQ) in settings where first and second derivatives may be evaluated along with the function itself. We perform sampling-based inference in order to incorporate uncertainty over hyperparameters, and show that both hyperparameter and function uncertainty decrease much more rapidly when using derivative information. Moreover, we introduce techniques for overcoming ill-conditioning issues that have plagued earlier methods for gradient-enhanced Gaussian processes and kriging. We illustrate the efficacy of these methods using applications to real and simulated Bayesian optimization and quadrature problems, and show that exploting derivatives can provide substantial gains over standard methods.

20 pages, 8 figures

References in corpus (2)

• Nested Variational Compression in Deep Gaussian Processes
• Sampling for Inference in Probabilistic Models with Fast Bayesian Quadrature

Cited by in corpus (9)

• Sobolev Training for Neural Networks
• Bayesian optimization with improved scalability and derivative information for efficient design of nanophotonic structures
• A Batched Scalable Multi-Objective Bayesian Optimization Algorithm
• Information Directed Sampling for Linear Partial Monitoring
• Learning Personalized Thermal Preferences via Bayesian Active Learning with Unimodality Constraints
• Using Distance Correlation for Efficient Bayesian Optimization
• Estimating a probability of failure with the convex order in computer experiments
• Stable Bayesian Optimisation via Direct Stability Quantification
• Scalable Bayesian Optimization with Sparse Gaussian Process Models