Tackling the XAI Disagreement Problem with Regional Explanations - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2024

Tackling the XAI Disagreement Problem with Regional Explanations

Résumé

The XAI Disagreement Problem concerns the fact that various explainability methods yield different local/global insights on model behavior. Thus, given the lack of ground truth in explainability, practitioners are left wondering "Which explanation should I believe?". In this work, we approach the Disagreement Problem from the point of view of Functional Decomposition (FD). First, we demonstrate that many XAI techniques disagree because they handle feature interactions differently. Secondly, we reduce interactions locally by fitting a so-called FD-Tree, which partitions the input space into regions where the model is approximately additive. Thus instead of providing global explanations aggregated over the whole dataset, we advocate reporting the FD-Tree structure as well as the regional explanations extracted from its leaves. The beneficial effects of FD-Trees on the Disagreement Problem are demonstrated on toy and real datasets.
Fichier principal
Vignette du fichier
main.pdf (2.15 Mo) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04480870 , version 1 (27-02-2024)

Licence

Identifiants

  • HAL Id : hal-04480870 , version 1

Citer

Gabriel Laberge, Yann Pequignot, Mario Marchand, Foutse Khomh. Tackling the XAI Disagreement Problem with Regional Explanations. International Conference on Artificial Intelligence and Statistics (AISTATS), May 2024, Valencia, Spain. ⟨hal-04480870⟩
169 Consultations
86 Téléchargements

Partager

Gmail Mastodon Facebook X LinkedIn More