The Distributional Uncertainty of the SHAP Score in Explainable Machine Learning

Santiago Cifuentes, Leopoldo Bertossi, Nina Pardal, Sergio Abriola, Maria Vanina Martinez, Miguel Romero

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Attribution scores reflect how important the feature values in an input entity are for the output of a machine learning model. One of the most popular attribution scores is the SHAP score, which is an instantiation of the general Shapley value used in coalition game theory. The definition of this score relies on a probability distribution on the entity population. Since the exact distribution is generally unknown, it needs to be assigned subjectively or be estimated from data, which may lead to misleading feature scores. In this paper, we propose a principled framework for reasoning on SHAP scores under unknown entity population distributions. In our framework, we consider an uncertainty region that contains the potential distributions, and the SHAP score of a feature becomes a function defined over this region. We study the basic problems of finding maxima and minima of this function, which allows us to determine tight ranges for the SHAP scores of all features. In particular, we pinpoint the complexity of these problems, and other related ones, showing them to be intractable. Finally, we present experiments on a real-world dataset, showing that our framework may contribute to a more robust feature scoring.
Original languageEnglish
Title of host publicationECAI 2024
Subtitle of host publication27th European Conference on Artificial Intelligence, 19–24 October 2024, Santiago de Compostela, Spain – Including 13th Conference on Prestigious Applications of Intelligent Systems (PAIS 2024)
EditorsUlle Endriss, Francisco S. Melo, Kerstin Bach, Alberto Bugarín-Diz, José M. Alonso-Moral, Senén Barro, Fredrik Heintz
PublisherIOS Press
Pages971-978
Number of pages8
Volume392
ISBN (Electronic)9781643685489
DOIs
Publication statusPublished - 19 Oct 2024
Externally publishedYes
Event27th European Conference on Artificial Intelligence - Including 13th Conference on Prestigious Applications of Intelligent Systems (PAIS 2024) - Santiago de Compostela, Spain
Duration: 19 Oct 202424 Oct 2024
Conference number: 27

Publication series

NameFrontiers in Artificial Intelligence and Applications
PublisherIOS Press
Number2024
Volume392
ISSN (Print)0922-6389
ISSN (Electronic)1879-8314

Conference

Conference27th European Conference on Artificial Intelligence - Including 13th Conference on Prestigious Applications of Intelligent Systems (PAIS 2024)
Abbreviated titleECAI 2024
Country/TerritorySpain
CitySantiago de Compostela
Period19/10/2424/10/24

Cite this