shrinkGPR: Scalable Gaussian Process Regression with Hierarchical Shrinkage Priors

Efficient variational inference methods for fully Bayesian Gaussian Process Regression (GPR) models with hierarchical shrinkage priors, including the triple gamma prior for effective variable selection and covariance shrinkage in high-dimensional settings. The package leverages normalizing flows to approximate complex posterior distributions. For details on implementation, see Knaus (2025) <doi:10.48550/arXiv.2501.13173>.

Version:	1.0.0
Depends:	R (≥ 4.0.0)
Imports:	gsl, progress, rlang, utils, methods, torch
Suggests:	testthat (≥ 3.0.0)
Published:	2025-01-30
DOI:	10.32614/CRAN.package.shrinkGPR
Author:	Peter Knaus [aut, cre]
Maintainer:	Peter Knaus <peter.knaus at wu.ac.at>
License:	GPL-2 | GPL-3 [expanded from: GPL (≥ 2)]
NeedsCompilation:	no
SystemRequirements:	torch backend, for installation guide see https://cran.r-project.org/web/packages/torch/vignettes/installation.html
CRAN checks:	shrinkGPR results

Documentation:

Reference manual:

shrinkGPR.pdf

Downloads:

Package source:	shrinkGPR_1.0.0.tar.gz
Windows binaries:	r-devel: shrinkGPR_1.0.0.zip, r-release: shrinkGPR_1.0.0.zip, r-oldrel: shrinkGPR_1.0.0.zip
macOS binaries:	r-devel (arm64): shrinkGPR_1.0.0.tgz, r-release (arm64): shrinkGPR_1.0.0.tgz, r-oldrel (arm64): shrinkGPR_1.0.0.tgz, r-devel (x86_64): shrinkGPR_1.0.0.tgz, r-release (x86_64): shrinkGPR_1.0.0.tgz, r-oldrel (x86_64): shrinkGPR_1.0.0.tgz

Linking:

Please use the canonical form https://CRAN.R-project.org/package=shrinkGPR to link to this page.