mgpr: An R package for multivariate Gaussian process regression
Gaussian process regression (GPR) is a non-parametric kernel-based machine learning method. GPR is based on Bayesian formalism, which enables the estimation of prediction uncertainty of the response variables. We propose an R package that provides an easy-to-use interface for multivariate GPR. The m...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Elsevier
2023-12-01
|
Series: | SoftwareX |
Subjects: | |
Online Access: | http://www.sciencedirect.com/science/article/pii/S2352711023002595 |
_version_ | 1797389472856276992 |
---|---|
author | Petri Varvia Janne Räty Petteri Packalen |
author_facet | Petri Varvia Janne Räty Petteri Packalen |
author_sort | Petri Varvia |
collection | DOAJ |
description | Gaussian process regression (GPR) is a non-parametric kernel-based machine learning method. GPR is based on Bayesian formalism, which enables the estimation of prediction uncertainty of the response variables. We propose an R package that provides an easy-to-use interface for multivariate GPR. The mgpr package was originally developed for remote sensing-based forest inventories that require multivariate prediction of forest attributes. The mgpr package supports both univariate and multivariate responses using a separable kernel and includes a robust hyperparameter estimation algorithm. The mgpr package is suitable for various regression problems with single response or multiple responses and provides good prediction performance. |
first_indexed | 2024-03-08T22:56:27Z |
format | Article |
id | doaj.art-62e9eaa72fe648a681b579ea54a44e88 |
institution | Directory Open Access Journal |
issn | 2352-7110 |
language | English |
last_indexed | 2024-03-08T22:56:27Z |
publishDate | 2023-12-01 |
publisher | Elsevier |
record_format | Article |
series | SoftwareX |
spelling | doaj.art-62e9eaa72fe648a681b579ea54a44e882023-12-16T06:08:11ZengElsevierSoftwareX2352-71102023-12-0124101563mgpr: An R package for multivariate Gaussian process regressionPetri Varvia0Janne Räty1Petteri Packalen2University of Eastern Finland, School of Forest Sciences, Yliopistokatu 2, 80100, Joensuu, Finland; Corresponding author.Natural Resources Institute Finland, Yliopistokatu 6 B, 80100 Joensuu, FinlandNatural Resources Institute Finland, Latokartanonkaari 9, 00790 Helsinki, Finland; University of Eastern Finland, School of Forest Sciences, Yliopistokatu 2, 80100, Joensuu, FinlandGaussian process regression (GPR) is a non-parametric kernel-based machine learning method. GPR is based on Bayesian formalism, which enables the estimation of prediction uncertainty of the response variables. We propose an R package that provides an easy-to-use interface for multivariate GPR. The mgpr package was originally developed for remote sensing-based forest inventories that require multivariate prediction of forest attributes. The mgpr package supports both univariate and multivariate responses using a separable kernel and includes a robust hyperparameter estimation algorithm. The mgpr package is suitable for various regression problems with single response or multiple responses and provides good prediction performance.http://www.sciencedirect.com/science/article/pii/S2352711023002595Bayesian approachCovariance functionStochastic process |
spellingShingle | Petri Varvia Janne Räty Petteri Packalen mgpr: An R package for multivariate Gaussian process regression SoftwareX Bayesian approach Covariance function Stochastic process |
title | mgpr: An R package for multivariate Gaussian process regression |
title_full | mgpr: An R package for multivariate Gaussian process regression |
title_fullStr | mgpr: An R package for multivariate Gaussian process regression |
title_full_unstemmed | mgpr: An R package for multivariate Gaussian process regression |
title_short | mgpr: An R package for multivariate Gaussian process regression |
title_sort | mgpr an r package for multivariate gaussian process regression |
topic | Bayesian approach Covariance function Stochastic process |
url | http://www.sciencedirect.com/science/article/pii/S2352711023002595 |
work_keys_str_mv | AT petrivarvia mgpranrpackageformultivariategaussianprocessregression AT janneraty mgpranrpackageformultivariategaussianprocessregression AT petteripackalen mgpranrpackageformultivariategaussianprocessregression |