Prediction Explanation with Dependence-Aware Shapley Values

Complex machine learning models are often hard to interpret. However, in many situations it is crucial to understand and explain why a model made a specific prediction. Shapley values is the only method for such prediction explanation framework with a solid theoretical foundation. Previously known methods for estimating the Shapley values do, however, assume feature independence. This package implements the method described in Aas, Jullum and Løland (2019) , which accounts for any feature dependence, and thereby produces more accurate estimates of the true Shapley values.


Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.


0.2.0 by Martin Jullum, a month ago,

Report a bug at

Browse source code at

Authors: Nikolai Sellereite [aut] , Martin Jullum [cre, aut] , Annabelle Redelmeier [aut] , Anders Løland [ctb] , Jens Christian Wahl [ctb] , Camilla Lingjærde [ctb] , Norsk Regnesentral [cph, fnd]

Documentation:   PDF Manual  

MIT + file LICENSE license

Imports stats, data.table, Rcpp, condMVNorm, mvnfast, Matrix

Suggests ranger, xgboost, mgcv, testthat, knitr, rmarkdown, roxygen2, MASS, ggplot2, caret, gbm, party, partykit

Linking to RcppArmadillo, Rcpp

See at CRAN