Näytä suppeat kuvailutiedot

dc.contributor.authorLindsten, Fredrik
dc.contributor.authorHelske, Jouni
dc.contributor.authorVihola, Matti
dc.contributor.editorBengio, S.
dc.contributor.editorWallach, H.
dc.contributor.editorLarochelle, H.
dc.contributor.editorGrauman, K.
dc.contributor.editorCesa-Bianchi, N.
dc.contributor.editorGarnett, R.
dc.date.accessioned2019-04-01T09:22:43Z
dc.date.available2019-04-01T09:22:43Z
dc.date.issued2018
dc.identifier.citationLindsten, F., Helske, J., & Vihola, M. (2018). Graphical model inference : Sequential Monte Carlo meets deterministic approximations. In S. Bengio, H. Wallach, H. Larochelle, K. Grauman, N. Cesa-Bianchi, & R. Garnett (Eds.), <i>NeurIPS 2018 : Proceedings of the 32nd Conference on Neural Information Processing Systems</i>. Neural Information Processing Systems Foundation, Inc.. Advances in Neural Information Processing Systems, 31. <a href="https://papers.nips.cc/paper/8041-graphical-model-inference-sequential-monte-carlo-meets-deterministic-approximations" target="_blank">https://papers.nips.cc/paper/8041-graphical-model-inference-sequential-monte-carlo-meets-deterministic-approximations</a>
dc.identifier.otherCONVID_28980157
dc.identifier.otherTUTKAID_81012
dc.identifier.urihttps://jyx.jyu.fi/handle/123456789/63322
dc.description.abstractApproximate inference in probabilistic graphical models (PGMs) can be grouped into deterministic methods and Monte-Carlo-based methods. The former can often provide accurate and rapid inferences, but are typically associated with biases that are hard to quantify. The latter enjoy asymptotic consistency, but can suffer from high computational costs. In this paper we present a way of bridging the gap between deterministic and stochastic inference. Specifically, we suggest an efficient sequential Monte Carlo (SMC) algorithm for PGMs which can leverage the output from deterministic inference methods. While generally applicable, we show explicitly how this can be done with loopy belief propagation, expectation propagation, and Laplace approximations. The resulting algorithm can be viewed as a post-correction of the biases associated with these methods and, indeed, numerical results show clear improvements over the baseline deterministic methods as well as over “plain” SMC.fi
dc.format.mimetypeapplication/pdf
dc.language.isoeng
dc.publisherNeural Information Processing Systems Foundation, Inc.
dc.relation.ispartofNeurIPS 2018 : Proceedings of the 32nd Conference on Neural Information Processing Systems
dc.relation.ispartofseriesAdvances in Neural Information Processing Systems
dc.relation.urihttps://papers.nips.cc/paper/8041-graphical-model-inference-sequential-monte-carlo-meets-deterministic-approximations
dc.rightsIn Copyright
dc.subject.otherkoneoppiminenfi
dc.subject.othertilastolliset mallitfi
dc.subject.othermachine learningfi
dc.subject.otherstatistical modelsfi
dc.titleGraphical model inference : Sequential Monte Carlo meets deterministic approximations
dc.typeconferenceObject
dc.identifier.urnURN:NBN:fi:jyu-201903261971
dc.contributor.laitosMatematiikan ja tilastotieteen laitosfi
dc.contributor.laitosDepartment of Mathematics and Statisticsen
dc.contributor.oppiaineTilastotiedefi
dc.contributor.oppiaineStatisticsen
dc.type.urihttp://purl.org/eprint/type/ConferencePaper
dc.date.updated2019-03-26T10:15:35Z
dc.type.coarhttp://purl.org/coar/resource_type/c_5794
dc.description.reviewstatuspeerReviewed
dc.relation.issn1049-5258
dc.relation.numberinseries31
dc.type.versionpublishedVersion
dc.rights.copyright© The Authors & NIPS, 2018.
dc.rights.accesslevelopenAccessfi
dc.relation.conferenceAdvances in neural information processing systems
dc.relation.grantnumber274740
dc.relation.grantnumber284513
dc.relation.grantnumber312605
dc.subject.ysotilastolliset mallit
dc.subject.ysokoneoppiminen
dc.format.contentfulltext
jyx.subject.urihttp://www.yso.fi/onto/yso/p26278
jyx.subject.urihttp://www.yso.fi/onto/yso/p21846
dc.rights.urlhttp://rightsstatements.org/page/InC/1.0/?language=en
dc.relation.funderSuomen Akatemiafi
dc.relation.funderSuomen Akatemiafi
dc.relation.funderSuomen Akatemiafi
dc.relation.funderAcademy of Finlanden
dc.relation.funderAcademy of Finlanden
dc.relation.funderAcademy of Finlanden
jyx.fundingprogramAkatemiatutkija, SAfi
jyx.fundingprogramAkatemiatutkijan tutkimuskulut, SAfi
jyx.fundingprogramAkatemiatutkijan tutkimuskulut, SAfi
jyx.fundingprogramAcademy Research Fellow, AoFen
jyx.fundingprogramResearch costs of Academy Research Fellow, AoFen
jyx.fundingprogramResearch costs of Academy Research Fellow, AoFen
jyx.fundinginformationFL has received support from the Swedish Foundation for Strategic Research (SSF) via the project Probabilistic Modeling and Inference for Machine Learning (contract number: ICA16-0015) and from the Swedish Research Council (VR) via the projects Learning of Large-Scale Probabilistic Dynamical Models (contract number: 2016-04278) and NewLEADS – New Directions in Learning Dynamical Systems (contract number: 621-2016-06079). JH and MV have received support from the Academy of Finland (grants 274740, 284513 and 312605).
dc.type.okmA4


Aineistoon kuuluvat tiedostot

Thumbnail

Aineisto kuuluu seuraaviin kokoelmiin

Näytä suppeat kuvailutiedot

In Copyright
Ellei muuten mainita, aineiston lisenssi on In Copyright