Moore, C.J., Chua, A.J.K., Berry, C.P.L. and Gair, J.R. (2016) Fast methods for training Gaussian processes on large datasets. Royal Society Open Science, 3(5), 160125. (doi: 10.1098/rsos.160125)
Text
218919.pdf - Published Version Available under License Creative Commons Attribution. 539kB |
Abstract
Gaussian process regression (GPR) is a non-parametric Bayesian technique for interpolating or fitting data. The main barrier to further uptake of this powerful tool rests in the computational costs associated with the matrices which arise when dealing with large datasets. Here, we derive some simple results which we have found useful for speeding up the learning stage in the GPR algorithm, and especially for performing Bayesian model comparison between different covariance functions. We apply our techniques to both synthetic and real data and quantify the speed-up relative to using nested sampling to numerically evaluate model evidences.
Item Type: | Articles |
---|---|
Status: | Published |
Refereed: | Yes |
Glasgow Author(s) Enlighten ID: | Gair, Dr Jonathan and Berry, Dr Christopher |
Authors: | Moore, C.J., Chua, A.J.K., Berry, C.P.L., and Gair, J.R. |
College/School: | College of Science and Engineering > School of Physics and Astronomy |
Journal Name: | Royal Society Open Science |
Publisher: | The Royal Society |
ISSN: | 2054-5703 |
ISSN (Online): | 2054-5703 |
Copyright Holders: | Copyright © 2016 The Authors |
First Published: | First published in 3(5):160125 |
Publisher Policy: | Reproduced under a Creative Commons licence |
Related URLs: |
University Staff: Request a correction | Enlighten Editors: Update this record