Efficient Robust Bayesian Optimization for Arbitrary Uncertain Inputs
This work addresses input uncertainty in Bayesian Optimization for applications like machining and execution noise, offering a robust solution with theoretical guarantees, though it is incremental as it builds on existing methods.
The paper tackles the problem of input uncertainty in Bayesian Optimization, which causes performance fluctuations, by introducing AIRBO, a robust algorithm that models arbitrary uncertain inputs using Gaussian Process with Maximum Mean Discrepancy and achieves state-of-the-art performance in experiments.
Bayesian Optimization (BO) is a sample-efficient optimization algorithm widely employed across various applications. In some challenging BO tasks, input uncertainty arises due to the inevitable randomness in the optimization process, such as machining errors, execution noise, or contextual variability. This uncertainty deviates the input from the intended value before evaluation, resulting in significant performance fluctuations in the final result. In this paper, we introduce a novel robust Bayesian Optimization algorithm, AIRBO, which can effectively identify a robust optimum that performs consistently well under arbitrary input uncertainty. Our method directly models the uncertain inputs of arbitrary distributions by empowering the Gaussian Process with the Maximum Mean Discrepancy (MMD) and further accelerates the posterior inference via Nystrom approximation. Rigorous theoretical regret bound is established under MMD estimation error and extensive experiments on synthetic functions and real problems demonstrate that our approach can handle various input uncertainties and achieve state-of-the-art performance.