Parallel Bayesian Global Optimization of Expensive Functions

Abstract

We consider parallel global optimization of derivative-free expensive-to-evaluate functions, and propose an efficient method based on stochastic approximation for implementing a conceptual Bayesian optimization algorithm proposed by Ginsbourger et al. (2007). At the heart of this algorithm is maximizing the information criterion called the "multi-points expected improvement'', or the q-EI. To accomplish this, we use infinitessimal perturbation analysis (IPA) to construct a stochastic gradient estimator and show that this estimator is unbiased. We also show that the stochastic gradient ascent algorithm using the constructed gradient estimator converges to a stationary point of the q-EI surface, and therefore, as the number of multiple starts of the gradient ascent algorithm and the number of steps for each start grow large, the one-step Bayes optimal set of points is recovered. We show in numerical experiments that our method for maximizing the q-EI is faster than methods based on closed-form evaluation using high-dimensional integration, when considering many parallel function evaluations, and is comparable in speed when considering few. We also show that the resulting one-step Bayes optimal algorithm for parallel global optimization finds high-quality solutions with fewer evaluations than a heuristic based on approximately maximizing the q-EI. A high-quality open source implementation of this algorithm is available in the open source Metrics Optimization Engine (MOE).

References in corpus (2)

• Practical Bayesian Optimization of Machine Learning Algorithms
• Bayesian optimization for materials design

Cited by in corpus (24)

• A Tutorial on Bayesian Optimization
• Differentiable Expected Hypervolume Improvement for Parallel Multi-Objective Bayesian Optimization
• Parallel and Distributed Thompson Sampling for Large-scale Accelerated Exploration of Chemical Space
• Parallel Bayesian Optimization of Multiple Noisy Objectives with Expected Hypervolume Improvement
• Black-Box Optimization with Local Generative Surrogates
• Asynchronous Parallel Bayesian Optimisation via Thompson Sampling
• Hyperparameter Learning via Distributional Transfer
• Efficient and Scalable Batch Bayesian Optimization Using K-Means
• Fast Efficient Hyperparameter Tuning for Policy Gradients
• Integrals over Gaussians under Linear Domain Constraints
• PARyOpt: A software for Parallel Asynchronous Remote Bayesian Optimization
• Sampling Acquisition Functions for Batch Bayesian Optimization
• Budgeted Batch Bayesian Optimization With Unknown Batch Sizes
• Weakly-supervised Multi-output Regression via Correlated Gaussian Processes
• Choosing a Suitable Acquisition Function for Batch Bayesian Optimization: Comparison of Serial and Monte Carlo Approaches
• Discretization-free Knowledge Gradient Methods for Bayesian Optimization
• BINOCULARS for Efficient, Nonmyopic Sequential Experimental Design
• Efficient Nonmyopic Bayesian Optimization via One-Shot Multi-Step Trees
• Lookahead Acquisition Functions for Finite-Horizon Time-Dependent Bayesian Optimization and Application to Quantum Optimal Control
• Efficient Phase Diagram Sampling by Active Learning
• Adaptive Simulation-based Training of AI Decision-makers using Bayesian Optimization
• Simple and Scalable Parallelized Bayesian Optimization
• Automatic Calibration of Dynamic and Heterogeneous Parameters in Agent-based Model
• Hybrid Repeat/Multi-point Sampling for Highly Volatile Objective Functions