The Need for Empirical Evaluation of Explanation Quality - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2022

The Need for Empirical Evaluation of Explanation Quality

Résumé

Prototype networks (Li et al. 2018) provide explanations to users using a prototype vector; that is, a vector learned by the network representing a "typical" observation. In this work, we propose an approach that identifies relevant features in the input space used by the Prototype network. We find however that empirical evaluation of explanation quality is difficult without ground truth explanations. We include a discussion about developing methods for generating explanations, identifying when one explanation method is preferable to another, and the complications that arise when measuring explanation quality.
Fichier principal
Vignette du fichier
AAAI_Workshop.pdf (342.42 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03591012 , version 1 (28-02-2022)

Identifiants

  • HAL Id : hal-03591012 , version 1

Citer

Nicholas Halliwell, Fabien Gandon, Freddy Lecue, Serena Villata. The Need for Empirical Evaluation of Explanation Quality. AAAI 2022 - Workshop on Explainable Agency in Artificial Intelligence, Feb 2022, Vancouver, Canada. ⟨hal-03591012⟩
107 Consultations
128 Téléchargements

Partager

Gmail Facebook X LinkedIn More