Passer à la navigation principale Passer à la recherche Passer au contenu principal

Fairwashing: the risk of rationalization

  • Ulrich Aïvodji
  • , Hiromi Arai
  • , Olivier Fortineau
  • , Sébastien Gambs
  • , Satoshi Hara
  • , Alain Tapp
  • Université du Québec à Montréal
  • RIKEN
  • Japan Science and Technology Agency
  • ENSTA
  • The University of Osaka
  • UdeM
  • Mila

Résultats de recherche: Contribution à un journal › Article de conférence › Revue par des pairs

60 Citations (Scopus)

Résumé

Black-box explanation is the problem of explaining how a machine learning model – whose internal logic is hidden to the auditor and generally complex – produces its outcomes. Current approaches for solving this problem include model explanation, outcome explanation as well as model inspection. While these techniques can be beneficial by providing interpretability, they can be used in a negative manner to perform fairwashing, which we define as promoting the false perception that a machine learning model respects some ethical values. In particular, we demonstrate that it is possible to systematically rationalize decisions taken by an unfair black-box model using the model explanation as well as the outcome explanation approaches with a given fairness metric. Our solution, LaundryML, is based on a regularized rule list enumeration algorithm whose objective is to search for fair rule lists approximating an unfair black-box model. We empirically evaluate our rationalization technique on black-box models trained on real-world datasets and show that one can obtain rule lists with high fidelity to the black-box model while being considerably less unfair at the same time.

Langue originaleAnglais
Pages (de - à)161-170
Nombre de pages10
RevueProceedings of Machine Learning Research
Volume97
ÉtatPublié - 2019
Modification externeOui
Evénement36th International Conference on Machine Learning, ICML 2019 - Long Beach, Etats-Unis
Durée: 9 juin 2019 → 15 juin 2019

Classification disciplinaire ASJC Scopus

  • Logiciel
  • Ingénierie de contrôle et des systèmes
  • Statistiques et probabilités
  • Intelligence artificielle

Empreinte digitale

Voici les principaux termes ou expressions associés à « Fairwashing: the risk of rationalization ». Ces libellés thématiques sont générés à partir du titre et du résumé de la publication. Ensemble, ils forment une empreinte digitale unique.

Citer cette ressource