TY - GEN
T1 - Efficiently Computable Safety Bounds for Gaussian Processes in Active Learning
AU - Tebbe, Jörn
AU - Zimmer, Christoph
AU - Steland, Ansgar
AU - Lange-Hegermann, Markus
AU - Mies, Fabian
PY - 2024
Y1 - 2024
N2 - Active learning of physical systems must commonly respect practical safety constraints, which restricts the exploration of the design space. Gaussian Processes (GPs) and their calibrated uncertainty estimations are widely used for this purpose. In many technical applications the design space is explored via continuous trajectories, along which the safety needs to be assessed. This is particularly challenging for strict safety requirements in GP methods, as it employs computationally expensive Monte-Carlo sampling of high quantiles. We address these challenges by providing provable safety bounds based on the adaptively sampled median of the supremum of the posterior GP. Our method significantly reduces the number of samples required for estimating high safety probabilities, resulting in faster evaluation without sacrificing accuracy and exploration speed. The effectiveness of our safe active learning approach is demonstrated through extensive simulations and validated using a real-world engine example.
AB - Active learning of physical systems must commonly respect practical safety constraints, which restricts the exploration of the design space. Gaussian Processes (GPs) and their calibrated uncertainty estimations are widely used for this purpose. In many technical applications the design space is explored via continuous trajectories, along which the safety needs to be assessed. This is particularly challenging for strict safety requirements in GP methods, as it employs computationally expensive Monte-Carlo sampling of high quantiles. We address these challenges by providing provable safety bounds based on the adaptively sampled median of the supremum of the posterior GP. Our method significantly reduces the number of samples required for estimating high safety probabilities, resulting in faster evaluation without sacrificing accuracy and exploration speed. The effectiveness of our safe active learning approach is demonstrated through extensive simulations and validated using a real-world engine example.
UR - http://www.scopus.com/inward/record.url?scp=85194147371&partnerID=8YFLogxK
UR - https://proceedings.mlr.press/v238/tebbe24a.html
M3 - Conference contribution
AN - SCOPUS:85194147371
VL - 238
T3 - Proceedings of Machine Learning Research
BT - Proceedings of the 27th International Conference on Artificial Intelligence and Statistics (AISTATS) 2024, Valencia, Spain
PB - PMLR
T2 - 27th International Conference on Artificial Intelligence and Statistics (AISTATS)
Y2 - 2 May 2024 through 4 May 2024
ER -