Logo des Repositoriums
 

Safe Active Learning for Time-Series Modeling with Gaussian Processes

Kurzanzeige
dc.contributor.authorZimmer, Christoph
dc.contributor.authorMeister, Mona
dc.contributor.authorNguyen-Tuong, Duy
dc.contributor.editorDavid, Klaus
dc.contributor.editorGeihs, Kurt
dc.contributor.editorLange, Martin
dc.contributor.editorStumme, Gerd
dc.date.accessioned2019-08-27T12:55:25Z
dc.date.available2019-08-27T12:55:25Z
dc.date.issued2019
dc.description.abstractLearning time-series models is useful for many applications, such as simulation and forecasting. In this study, we consider the problem of actively learning time-series models while taking given safety constraints into account. For time-series modeling we employ a Gaussian process with a nonlinear exogenous input structure. The proposed approach generates data appropriate for time series model learning, i.e. input and output trajectories, by dynamically exploring the input space. The approach parametrizes the input trajectory as consecutive trajectory sections, which are determined stepwise given safety requirements and past observations. We analyze the proposed algorithm and evaluate it empirically on a technical application. The results show the effectiveness of our approach in a realistic technical use case.en
dc.identifier.doi10.18420/inf2019_44
dc.identifier.isbn978-3-88579-688-6
dc.identifier.pissn1617-5468
dc.identifier.urihttps://dl.gi.de/handle/20.500.12116/24993
dc.language.isoen
dc.publisherGesellschaft für Informatik e.V.
dc.relation.ispartofINFORMATIK 2019: 50 Jahre Gesellschaft für Informatik – Informatik für Gesellschaft
dc.relation.ispartofseriesLecture Notes in Informatics (LNI) - Proceedings, Volume P-294
dc.subjectSafe Active Learning
dc.subjectDynamics Modeling
dc.titleSafe Active Learning for Time-Series Modeling with Gaussian Processesen
dc.typeText/Conference Paper
gi.citation.endPage281
gi.citation.publisherPlaceBonn
gi.citation.startPage281
gi.conference.date23.-26. September 2019
gi.conference.locationKassel
gi.conference.sessiontitleData Science

Dateien

Originalbündel
1 - 1 von 1
Vorschaubild
Name:
paper3_21.pdf
Größe:
35.12 KB
Format:
Adobe Portable Document Format
Herunterladen

Sammlungen

P294 - INFORMATIK 2019 - 50 Jahre Gesellschaft für Informatik – Informatik für Gesellschaft