Debiasing SHAP scores in random forests

Markus Loecher ()
Additional contact information
Markus Loecher: Berlin School of Economics and Law

AStA Advances in Statistical Analysis, 2024, vol. 108, issue 2, No 9, 427-440

Abstract: Abstract Black box machine learning models are currently being used for high-stakes decision making in various parts of society such as healthcare and criminal justice. While tree-based ensemble methods such as random forests typically outperform deep learning models on tabular data sets, their built-in variable importance algorithms are known to be strongly biased toward high-entropy features. It was recently shown that the increasingly popular SHAP (SHapley Additive exPlanations) values suffer from a similar bias. We propose debiased or "shrunk" SHAP scores based on sample splitting which additionally enable the detection of overfitting issues at the feature level.

Keywords: Interpretable machine learning; Feature importance; Random forests; SHAP values; Explainable artificial intelligence (search for similar items in EconPapers)
Date: 2024
References: View complete reference list from CitEc
Citations:

Downloads: (external link)
http://link.springer.com/10.1007/s10182-023-00479-7 Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:alstar:v:108:y:2024:i:2:d:10.1007_s10182-023-00479-7

Ordering information: This journal article can be ordered from
http://www.springer. ... cs/journal/10182/PS2

DOI: 10.1007/s10182-023-00479-7

Access Statistics for this article

AStA Advances in Statistical Analysis is currently edited by Göran Kauermann and Yarema Okhrin

More articles in AStA Advances in Statistical Analysis from Springer, German Statistical Society
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().