Massively Parallel Approximate Gaussian Process Regression

dc.contributor.author Gramacy, Robert
dc.contributor.author Niemi, Jarad
dc.contributor.author Niemi, Jarad
dc.contributor.author Weiss, Robin
dc.contributor.department Statistics
dc.date 2018-02-18T05:20:31.000
dc.date.accessioned 2020-07-02T06:58:10Z
dc.date.available 2020-07-02T06:58:10Z
dc.date.issued 2014-09-30
dc.description.abstract <p>We explore how the big-three computing paradigms---symmetric multiprocessor, graphical processing units (GPUs), and cluster computing---can together be brought to bear on large-data Gaussian processes (GP) regression problems via a careful implementation of a newly developed local approximation scheme. Our methodological contribution focuses primarily on GPU computation, as this requires the most care and also provides the largest performance boost. However, in our empirical work we study the relative merits of all three paradigms to determine how best to combine them. The paper concludes with two case studies. One is a real data fluid-dynamics computer experiment which benefits from the local nature of our approximation; the second is a synthetic example designed to find the largest data set for which (accurate) GP emulation can be performed on a commensurate predictive set in under an hour.<br /><br /></p>
dc.description.comments <p>This is an article from <em>SIAM/ASA Journal on Uncertainty Quantification</em> 2 (2014): 564, <a href="http://dx.doi.org/10.1137/130941912" target="_blank">doi: 10.1137/130941912</a>. Posted with permission.</p>
dc.format.mimetype application/pdf
dc.identifier archive/lib.dr.iastate.edu/stat_las_pubs/91/
dc.identifier.articleid 1090
dc.identifier.contextkey 9817714
dc.identifier.s3bucket isulib-bepress-aws-west
dc.identifier.submissionpath stat_las_pubs/91
dc.identifier.uri https://dr.lib.iastate.edu/handle/20.500.12876/90694
dc.language.iso en
dc.source.bitstream archive/lib.dr.iastate.edu/stat_las_pubs/91/2014_Niemi_MassivelyParallel.pdf|||Sat Jan 15 02:28:24 UTC 2022
dc.source.uri 10.1137/130941912
dc.subject.disciplines Applied Statistics
dc.subject.disciplines Design of Experiments and Sample Surveys
dc.subject.disciplines Statistical Models
dc.subject.keywords emulator
dc.subject.keywords nonparametric regression
dc.subject.keywords graphical processing unit
dc.subject.keywords symmetric multiprocessor
dc.subject.keywords cluster computing
dc.subject.keywords big data
dc.subject.keywords computer experiment
dc.title Massively Parallel Approximate Gaussian Process Regression
dc.type article
dc.type.genre article
dspace.entity.type Publication
relation.isAuthorOfPublication 31b412ec-d498-4926-901e-2cb5c2b5a31d
relation.isOrgUnitOfPublication 264904d9-9e66-4169-8e11-034e537ddbca
File
Original bundle
Now showing 1 - 1 of 1
Name:
2014_Niemi_MassivelyParallel.pdf
Size:
314.42 KB
Format:
Adobe Portable Document Format
Description:
Collections