KRLS: Kernel-Based Regularized Least Squares

Implements Kernel-based Regularized Least Squares (KRLS), a machine learning method to fit multidimensional functions y = f(x) for regression and classification problems without relying on linearity or additivity assumptions. KRLS finds the best fitting function by minimizing the squared loss of a Tikhonov regularization problem, using Gaussian kernels as radial basis functions. For further details see Hainmueller and Hazlett (2014, <doi:10.1093/pan/mpt019>).

Version:	1.7-0
Imports:	grDevices, graphics, stats
Suggests:	lattice, testthat (≥ 3.0.0), knitr, rmarkdown, ggplot2, generics
Published:	2026-06-05
DOI:	10.32614/CRAN.package.KRLS
Author:	Jens Hainmueller [aut, cre], Chad Hazlett [aut]
Maintainer:	Jens Hainmueller <jhain at stanford.edu>
BugReports:	https://github.com/j-hai/KRLS/issues
License:	GPL-2 | GPL-3 [expanded from: GPL (≥ 2)]
URL:	https://web.stanford.edu/~jhain/, https://github.com/j-hai/KRLS
NeedsCompilation:	no
Citation:	KRLS citation info
Materials:	NEWS
CRAN checks:	KRLS results

Documentation:

Reference manual:	KRLS.html , KRLS.pdf
Vignettes:	Scaling KRLS to larger n: the Nystrom approximation (source, R code) Quickstart: kernel-regularized least squares with KRLS (source, R code)

Downloads:

Package source:	KRLS_1.7-0.tar.gz
Windows binaries:	r-devel: KRLS_1.7-0.zip, r-release: KRLS_1.7-0.zip, r-oldrel: KRLS_1.7-0.zip
macOS binaries:	r-release (arm64): KRLS_1.7-0.tgz, r-oldrel (arm64): KRLS_1.7-0.tgz, r-release (x86_64): KRLS_1.7-0.tgz, r-oldrel (x86_64): KRLS_1.7-0.tgz
Old sources:	KRLS archive

Reverse dependencies:

Reverse imports:

InfluenceBorrowing, qqkrls

Linking:

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