Learning Fuzzy Relations and Properties for Explainable Artificial Intelligence - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2018

Learning Fuzzy Relations and Properties for Explainable Artificial Intelligence

Résumé

The goal of explainable artificial intelligence is to solve problems in a way that humans can understand how it does it. However, few approaches have been proposed so far and some of them lay more emphasis on interpretabil-ity than on explainability. In this paper, we propose an approach that is based on learning fuzzy relations and fuzzy properties. We extract frequent relations from a dataset to generate an explained decision. Our approach can deal with different problems, such as classification or annotation. A model was built to perform explained classification on a toy dataset that we generated. It managed to correctly classify examples while providing convincing explanations. A few areas for improvement have been spotted, such as the need to filter relations and properties before or while learning them in order to avoid useless computations.
Fichier principal
Vignette du fichier
articleRio.pdf (582.92 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02425453 , version 1 (30-12-2019)

Identifiants

Citer

Régis Pierrard, Jean-Philippe Poli, Céline Hudelot. Learning Fuzzy Relations and Properties for Explainable Artificial Intelligence. 2018 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE), Jul 2018, Rio de Janeiro, Brazil. ⟨10.1109/FUZZ-IEEE.2018.8491538⟩. ⟨hal-02425453⟩
80 Consultations
666 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More