An implementation of the fair data adaptation with quantile
preservation described in Plecko & Meinshausen (2019) <arXiv:1911.06685>.
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.
Version: |
0.2.4 |
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 |
Published: |
2022-07-01 |
Author: |
Drago Plecko [aut, cre],
Nicolas Bennett [aut] |
Maintainer: |
Drago Plecko <drago.plecko at stat.math.ethz.ch> |
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 |