Scalable Global Optimization via Local Bayesian Optimization

TL;DR

This paper introduces TuRBO, a local Bayesian optimization algorithm that efficiently handles high-dimensional, large-scale black-box problems by using multiple local models and a bandit-based sampling strategy, outperforming existing methods.

Contribution

The paper proposes TuRBO, a novel local Bayesian optimization method that improves scalability and performance on high-dimensional problems by combining local models with a bandit-based sample allocation.

Findings

TuRBO outperforms state-of-the-art optimization methods.

Effective for high-dimensional, large-scale problems.

Applicable across diverse domains like reinforcement learning and robotics.

Abstract

Bayesian optimization has recently emerged as a popular method for the sample-efficient optimization of expensive black-box functions. However, the application to high-dimensional problems with several thousand observations remains challenging, and on difficult problems Bayesian optimization is often not competitive with other paradigms. In this paper we take the view that this is due to the implicit homogeneity of the global probabilistic models and an overemphasized exploration that results from global acquisition. This motivates the design of a local probabilistic approach for global optimization of large-scale high-dimensional problems. We propose the $\texttt{TuRBO}$ algorithm that fits a collection of local models and performs a principled global allocation of samples across these models via an implicit bandit approach. A comprehensive evaluation demonstrates that $\texttt{TuRBO}$ outperforms state-of-the-art methods from machine learning and operations research on problems spanning reinforcement learning, robotics, and the natural sciences.