Robust Entropy Search for Safe Efficient Bayesian Optimization
Machine Learning
2024-06-03 v2 Machine Learning
Abstract
The practical use of Bayesian Optimization (BO) in engineering applications imposes special requirements: high sampling efficiency on the one hand and finding a robust solution on the other hand. We address the case of adversarial robustness, where all parameters are controllable during the optimization process, but a subset of them is uncontrollable or even adversely perturbed at the time of application. To this end, we develop an efficient information-based acquisition function that we call Robust Entropy Search (RES). We empirically demonstrate its benefits in experiments on synthetic and real-life data. The results showthat RES reliably finds robust optima, outperforming state-of-the-art algorithms.
Cite
@article{arxiv.2405.19059,
title = {Robust Entropy Search for Safe Efficient Bayesian Optimization},
author = {Dorina Weichert and Alexander Kister and Sebastian Houben and Patrick Link and Gunar Ernis},
journal= {arXiv preprint arXiv:2405.19059},
year = {2024}
}