Theoretical Evaluation of Asymmetric Shapley Values for Root-Cause Analysis

Kelen, Domokos; Petreczky, Mihàly; Kersch, Péter; Benczúr, Andràs

Type de document :

Communication dans un congrès avec actes

DOI :

10.1109/ICDM58522.2023.00030

Titre :

Theoretical Evaluation of Asymmetric Shapley Values for Root-Cause Analysis

Auteur(s) :

Kelen, Domokos [Auteur]
Institute for Computer Science and Control [Budapest] [SZTAKI]
Petreczky, Mihàly [Auteur]
Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 [CRIStAL]
Kersch, Péter [Auteur]
Ericsson [Budapest]
Benczúr, Andràs [Auteur]
Institute for Computer Science and Control [Budapest] [SZTAKI]

Titre de la manifestation scientifique :

2023 IEEE International Conference on Data Mining (ICDM)

Ville :

Shanghai

Pays :

Chine

Date de début de la manifestation scientifique :

2023-12-01

Titre de l’ouvrage :

2023 IEEE International Conference on Data Mining (ICDM)

Éditeur :

IEEE

Mot(s)-clé(s) en anglais :

explainability SHAP causality
explainability
SHAP
causality

Discipline(s) HAL :

Informatique [cs]/Intelligence artificielle [cs.AI]

Résumé en anglais : [en]

In this work, we examine Asymmetric Shapley Values (ASV), a variant of the popular SHAP additive local explanation method. ASV proposes a way to improve model explanations incorporating known causal relations between ...
Lire la suite >In this work, we examine Asymmetric Shapley Values (ASV), a variant of the popular SHAP additive local explanation method. ASV proposes a way to improve model explanations incorporating known causal relations between variables, and is also considered as a way to test for unfair discrimination in model predictions. Unexplored in previous literature, relaxing symmetry in Shapley values can have counter-intuitive consequences for model explanation. To better understand the method, we first show how local contributions correspond to global contributions of variance reduction. Using variance, we demonstrate multiple cases where ASV yields counter-intuitive attributions, arguably producing incorrect results for root-cause analysis. Second, we identify generalized additive models (GAM) as a restricted class for which ASV exhibits desirable properties. We support our arguments by proving multiple theoretical results about the method. Finally, we demonstrate the use of asymmetric attributions on multiple real-world datasets, comparing the results with and without restricted model families using gradient boosting and deep learning models.Lire moins >

Langue :

Anglais

Comité de lecture :

Oui

Audience :

Internationale

Vulgarisation :

Non

Collections :

Centre de Recherche en Informatique, Signal et Automatique de Lille (CRIStAL) - UMR 9189

Source :

Harvested from HAL

Fichiers

document
Accès libre
Accéder au document

2310.09961.pdf
Accès libre
Accéder au document

document
Accès libre
Accéder au document

2310.09961.pdf
Accès libre
Accéder au document

Theoretical Evaluation of Asymmetric Shapley ... BibTeX CSV Excel RIS

Fichiers

Theoretical Evaluation of Asymmetric Shapley ...

BibTeX

CSV

Excel

RIS