Efficiently Computable Safety Bounds for Gaussian Processes in Active Learning

Jörn Tebbe, Christoph Zimmer, Ansgar Steland, Markus Lange-Hegermann, Fabian Mies

Research output: Chapter in Book/Conference proceedings/Edited volumeConference contributionScientificpeer-review

20 Downloads (Pure)

Abstract

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.
Original languageEnglish
Title of host publicationProceedings of the 27th International Conference on Artificial Intelligence and Statistics (AISTATS) 2024, Valencia, Spain
PublisherPMLR
Number of pages21
Volume238
Publication statusPublished - 2024
Event27th International Conference on Artificial Intelligence and Statistics (AISTATS) - Valencia, Spain
Duration: 2 May 20244 May 2024

Publication series

NameProceedings of Machine Learning Research
ISSN (Print)2640-3498

Conference

Conference27th International Conference on Artificial Intelligence and Statistics (AISTATS)
Country/TerritorySpain
CityValencia
Period2/05/244/05/24

Fingerprint

Dive into the research topics of 'Efficiently Computable Safety Bounds for Gaussian Processes in Active Learning'. Together they form a unique fingerprint.

Cite this