Increasing the Scope as You Learn: Adaptive Bayesian Optimization in Nested Subspaces
This addresses a critical bottleneck for applications in life sciences, neural architecture search, and robotics by improving reliability and performance in high-dimensional optimization.
The paper tackles the problem of degrading performance and failure risks in high-dimensional Bayesian optimization (HDBO) by proposing BAxUS, which uses nested random subspaces to adaptively optimize, achieving better results than state-of-the-art methods across a broad set of applications.
Recent advances have extended the scope of Bayesian optimization (BO) to expensive-to-evaluate black-box functions with dozens of dimensions, aspiring to unlock impactful applications, for example, in the life sciences, neural architecture search, and robotics. However, a closer examination reveals that the state-of-the-art methods for high-dimensional Bayesian optimization (HDBO) suffer from degrading performance as the number of dimensions increases or even risk failure if certain unverifiable assumptions are not met. This paper proposes BAxUS that leverages a novel family of nested random subspaces to adapt the space it optimizes over to the problem. This ensures high performance while removing the risk of failure, which we assert via theoretical guarantees. A comprehensive evaluation demonstrates that BAxUS achieves better results than the state-of-the-art methods for a broad set of applications.