fairadapt: Fair Data Adaptation with Quantile Preservation

An implementation of the fair data adaptation with quantile preservation described in Plecko & Meinshausen (JMLR 2020, 21(242), 1-44). The adaptation procedure uses the specified causal graph to pre-process the given training and testing data in such a way to remove the bias caused by the protected attribute. The procedure uses tree ensembles for quantile regression. Instructions for using the methods are further elaborated in the corresponding JSS manuscript, see <doi:10.18637/jss.v110.i04>.

Version: 1.0.0
Depends: R (≥ 3.5.0)
Imports: ranger (≥ 0.13.1), assertthat, quantreg, qrnn, igraph, ggplot2, cowplot, scales
Suggests: testthat (≥ 3.0.3), knitr, rmarkdown, rticles, mvtnorm, magick, ggraph, pdftools, microbenchmark, xtable, spelling
Published: 2024-09-06
DOI: 10.32614/CRAN.package.fairadapt
Author: Drago Plecko [aut, cre], Nicolas Bennett [aut]
Maintainer: Drago Plecko <www.plecko at gmail.com>
BugReports: https://github.com/dplecko/fairadapt/issues
License: GPL (≥ 3)
URL: https://github.com/dplecko/fairadapt
NeedsCompilation: no
Language: en-US
Citation: fairadapt citation info
Materials: README NEWS
CRAN checks: fairadapt results

Documentation:

Reference manual: fairadapt.pdf
Vignettes: fairadapt: Causal Reasoning for Fair Data Preprocessing (source, R code)

Downloads:

Package source: fairadapt_1.0.0.tar.gz
Windows binaries: r-devel: fairadapt_1.0.0.zip, r-release: fairadapt_1.0.0.zip, r-oldrel: fairadapt_1.0.0.zip
macOS binaries: r-release (arm64): fairadapt_0.2.7.tgz, r-oldrel (arm64): not available, r-release (x86_64): fairadapt_1.0.0.tgz, r-oldrel (x86_64): fairadapt_1.0.0.tgz
Old sources: fairadapt archive

Linking:

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