Published by the Foundation for Open Access Statistics Editors-in-chief: Bettina Grün, Torsten Hothorn, Rebecca Killick, Edzer Pebesma, Achim Zeileis    ISSN 1548-7660; CODEN JSSOBK
Authors: Robert B. Gramacy
Title: laGP: Large-Scale Spatial Modeling via Local Approximate Gaussian Processes in R
Abstract: Gaussian process (GP) regression models make for powerful predictors in out of sample exercises, but cubic runtimes for dense matrix decompositions severely limit the size of data - training and testing - on which they can be deployed. That means that in computer experiment, spatial/geo-physical, and machine learning contexts, GPs no longer enjoy privileged status as data sets continue to balloon in size. We discuss an implementation of local approximate Gaussian process models, in the laGP package for R, that offers a particular sparse-matrix remedy uniquely positioned to leverage modern parallel computing architectures. The laGP approach can be seen as an update on the spatial statistical method of local kriging neighborhoods. We briefly review the method, and provide extensive illustrations of the features in the package through worked-code examples. The appendix covers custom building options for symmetric multi-processor and graphical processing units, and built-in wrapper routines that automate distribution over a simple network of workstations.

Page views:: 3028. Submitted: 2014-08-04. Published: 2016-08-16.
Paper: laGP: Large-Scale Spatial Modeling via Local Approximate Gaussian Processes in R     Download PDF (Downloads: 1869)
laGP_1.3.tar.gz: R source package Download (Downloads: 128; 1MB)
v72i01.R: R replication code Download (Downloads: 188; 18KB)

DOI: 10.18637/jss.v072.i01

This work is licensed under the licenses
Paper: Creative Commons Attribution 3.0 Unported License
Code: GNU General Public License (at least one of version 2 or version 3) or a GPL-compatible license.