Weichert, DorinaDorinaWeichertKister, AlexanderAlexanderKisterHouben, SebastianSebastianHoubenLink, PatrickPatrickLinkErnis, GunarGunarErnis2024-11-142025-01-142024-11-142024-09https://publica.fraunhofer.de/handle/publica/478940The 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 show that RES reliably finds robust optima, outperforming state-of-the-art algorithms.enrobustnessBayesian Optimizationrobust optimizationconference paper