DC Field | Value | Language |
---|---|---|
dc.contributor.author | Jeon, Younghwan | ko |
dc.contributor.author | Hwang, Ganguk | ko |
dc.date.accessioned | 2022-04-25T06:00:07Z | - |
dc.date.available | 2022-04-25T06:00:07Z | - |
dc.date.created | 2022-04-25 | - |
dc.date.issued | 2022-07 | - |
dc.identifier.citation | PATTERN RECOGNITION, v.127 | - |
dc.identifier.issn | 0031-3203 | - |
dc.identifier.uri | http://hdl.handle.net/10203/295865 | - |
dc.description.abstract | We address the data association problem and propose a Bayesian approach based on a mixture of Gaus-sian Processes (GPs) having two key components, the assignment probabilities and the GPs. In the pro-posed approach, the two key components are simultaneously updated according to observations through an efficient Expectation-Maximization (EM) algorithm that we develop. The proposed approach is thus more adaptive to the observations than the existing approaches for data association. To validate the per-formance of the proposed approach, we provide experimental results with real data sets as well as two synthetic data sets. We also provide a theoretical analysis to show the effectiveness of the Bayesian up -date.(c) 2022 Elsevier Ltd. All rights reserved. | - |
dc.language | English | - |
dc.publisher | ELSEVIER SCI LTD | - |
dc.title | Bayesian mixture of gaussian processes for data association problem | - |
dc.type | Article | - |
dc.identifier.wosid | 000776971700008 | - |
dc.identifier.scopusid | 2-s2.0-85125255489 | - |
dc.type.rims | ART | - |
dc.citation.volume | 127 | - |
dc.citation.publicationname | PATTERN RECOGNITION | - |
dc.identifier.doi | 10.1016/j.patcog.2022.108592 | - |
dc.contributor.localauthor | Hwang, Ganguk | - |
dc.description.isOpenAccess | N | - |
dc.type.journalArticle | Article | - |
dc.subject.keywordAuthor | Gaussian processes | - |
dc.subject.keywordAuthor | Bayesian models | - |
dc.subject.keywordAuthor | Variational inference | - |
dc.subject.keywordAuthor | Expectation maximization | - |
dc.subject.keywordPlus | MODEL | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.