mcboost: Multi-Calibration Boosting

Implements 'Multi-Calibration Boosting' (2018) <> and 'Multi-Accuracy Boosting' (2019) <arXiv:1805.12317> for the multi-calibration of a machine learning model's prediction. 'MCBoost' updates predictions for sub-groups in an iterative fashion in order to mitigate biases like poor calibration or large accuracy differences across subgroups. Multi-Calibration works best in scenarios where the underlying data & labels are unbiased, but resulting models are. This is often the case, e.g. when an algorithm fits a majority population while ignoring or under-fitting minority populations.

Depends: R (≥ 3.1.0)
Imports: backports, checkmate (≥ 2.0.0), data.table (≥ 1.13.6), mlr3 (≥ 0.10), mlr3misc (≥ 0.8.0), mlr3pipelines (≥ 0.3.0), R6 (≥ 2.4.1), rpart, glmnet
Suggests: curl, formattable, tidyverse, PracTools, mlr3learners, mlr3oml, neuralnet, paradox, testthat, knitr, ranger, rmarkdown, covr
Published: 2021-08-03
Author: Florian Pfisterer ORCID iD [cre, aut], Susanne Dandl ORCID iD [ctb], Christoph Kern ORCID iD [ctb], Bernd Bischl ORCID iD [ctb]
Maintainer: Florian Pfisterer <pfistererf at>
License: LGPL (≥ 3)
NeedsCompilation: no
Materials: README NEWS
CRAN checks: mcboost results


Reference manual: mcboost.pdf
Vignettes: MCBoost - Basics and Extensions
MCBoost - Health Survey Example


Package source: mcboost_0.3.3.0.tar.gz
Windows binaries: r-devel:, r-release:, r-oldrel:
macOS binaries: r-release (arm64): mcboost_0.3.3.0.tgz, r-release (x86_64): mcboost_0.3.3.0.tgz, r-oldrel: mcboost_0.3.3.0.tgz
Old sources: mcboost archive


Please use the canonical form to link to this page.