Campus Units
Statistics
Document Type
Article
Publication Version
Published Version
Publication Date
9-30-2014
Journal or Book Title
SIAM/ASA Journal on Uncertainty Quantification
Volume
2
Issue
1
First Page
564
Last Page
584
DOI
10.1137/130941912
Abstract
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.
Copyright Owner
2014
Copyright Date
Society for Industrial and Applied Mathematics
Language
en
File Format
application/pdf
Recommended Citation
Gramacy, Robert B.; Niemi, Jarad; and Weiss, Robin M., "Massively Parallel Approximate Gaussian Process Regression" (2014). Statistics Publications. 91.
https://lib.dr.iastate.edu/stat_las_pubs/91
Included in
Applied Statistics Commons, Design of Experiments and Sample Surveys Commons, Statistical Models Commons
Comments
This is an article from SIAM/ASA Journal on Uncertainty Quantification 2 (2014): 564, doi: 10.1137/130941912. Posted with permission.