2026-07-21T03:15:32Z https://drops.dagstuhl.de/oai

oai:drops-oai.dagstuhl.de:18032 2024-03-06T11:01:16Z ddc:004 open_access

The Safe and Effective Use of Low-Assurance Predictions in Safety-Critical Systems Agrawal, Kunal Baruah, Sanjoy Bender, Michael A. Marchetti-Spaccamela, Alberto Algorithms using predictions robust scheduling energy minimization classification on-line scheduling The algorithm-design paradigm of algorithms using predictions is explored as a means of incorporating the computations of lower-assurance components (such as machine-learning based ones) into safety-critical systems that must have their correctness validated to very high levels of assurance. The paradigm is applied to two simple example applications that are relevant to the real-time systems community: energy-aware scheduling, and classification using ML-based classifiers in conjunction with more reliable but slower deterministic classifiers. It is shown how algorithms using predictions achieve much-improved performance when the low-assurance computations are correct, at a cost of no more than a slight performance degradation even when they turn out to be completely wrong. Schloss Dagstuhl – Leibniz-Zentrum für Informatik Kunal Agrawal and Sanjoy Baruah and Michael A. Bender and Alberto Marchetti-Spaccamela 2023 Is Part Of LIPIcs, Volume 262, 35th Euromicro Conference on Real-Time Systems (ECRTS 2023) InProceedings Text doc-type:ResearchArticle publishedVersion application/pdf doi:10.4230/LIPIcs.ECRTS.2023.3 urn:nbn:de:0030-drops-180323 https://drops.dagstuhl.de/entities/document/10.4230/LIPIcs.ECRTS.2023.3 eng https://creativecommons.org/licenses/by/4.0/legalcode