Saved in:
Bibliographic Details
Main Authors: Marques-Silva, Joao, Huang, Xuanxiang, Letoffe, Olivier
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2501.11429
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913692130476032
author Marques-Silva, Joao
Huang, Xuanxiang
Letoffe, Olivier
author_facet Marques-Silva, Joao
Huang, Xuanxiang
Letoffe, Olivier
contents Recent work demonstrated the existence of critical flaws in the current use of Shapley values in explainable AI (XAI), i.e. the so-called SHAP scores. These flaws are significant in that the scores provided to a human decision-maker can be misleading. Although these negative results might appear to indicate that Shapley values ought not be used in XAI, this paper argues otherwise. Concretely, this paper proposes a novel definition of SHAP scores that overcomes existing flaws. Furthermore, the paper outlines a practically efficient solution for the rigorous estimation of the novel SHAP scores. Preliminary experimental results confirm our claims, and further underscore the flaws of the current SHAP scores.
format Preprint
id arxiv_https___arxiv_org_abs_2501_11429
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle The Explanation Game -- Rekindled (Extended Version)
Marques-Silva, Joao
Huang, Xuanxiang
Letoffe, Olivier
Artificial Intelligence
Recent work demonstrated the existence of critical flaws in the current use of Shapley values in explainable AI (XAI), i.e. the so-called SHAP scores. These flaws are significant in that the scores provided to a human decision-maker can be misleading. Although these negative results might appear to indicate that Shapley values ought not be used in XAI, this paper argues otherwise. Concretely, this paper proposes a novel definition of SHAP scores that overcomes existing flaws. Furthermore, the paper outlines a practically efficient solution for the rigorous estimation of the novel SHAP scores. Preliminary experimental results confirm our claims, and further underscore the flaws of the current SHAP scores.
title The Explanation Game -- Rekindled (Extended Version)
topic Artificial Intelligence
url https://arxiv.org/abs/2501.11429