Robust Entropy Search for Safe Efficient Bayesian Optimization

Dorina Weichert, Alexander Kister, Sebastian Houben, Patrick Link, Gunar Ernis
Proceedings of the Fortieth Conference on Uncertainty in Artificial Intelligence, PMLR 244:3711-3729, 2024.

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 show that RES reliably finds robust optima, outperforming state-of-the-art algorithms.

Cite this Paper

BibTeX
@InProceedings{pmlr-v244-weichert24a, title = {Robust Entropy Search for Safe Efficient Bayesian Optimization}, author = {Weichert, Dorina and Kister, Alexander and Houben, Sebastian and Link, Patrick and Ernis, Gunar}, booktitle = {Proceedings of the Fortieth Conference on Uncertainty in Artificial Intelligence}, pages = {3711--3729}, year = {2024}, editor = {Kiyavash, Negar and Mooij, Joris M.}, volume = {244}, series = {Proceedings of Machine Learning Research}, month = {15--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v244/main/assets/weichert24a/weichert24a.pdf}, url = {https://proceedings.mlr.press/v244/weichert24a.html}, 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 show that RES reliably finds robust optima, outperforming state-of-the-art algorithms.} }
Endnote
%0 Conference Paper %T Robust Entropy Search for Safe Efficient Bayesian Optimization %A Dorina Weichert %A Alexander Kister %A Sebastian Houben %A Patrick Link %A Gunar Ernis %B Proceedings of the Fortieth Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2024 %E Negar Kiyavash %E Joris M. Mooij %F pmlr-v244-weichert24a %I PMLR %P 3711--3729 %U https://proceedings.mlr.press/v244/weichert24a.html %V 244 %X 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 show that RES reliably finds robust optima, outperforming state-of-the-art algorithms.
APA
Weichert, D., Kister, A., Houben, S., Link, P. & Ernis, G.. (2024). Robust Entropy Search for Safe Efficient Bayesian Optimization. Proceedings of the Fortieth Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research 244:3711-3729 Available from https://proceedings.mlr.press/v244/weichert24a.html.

Related Material