Fairness in generative modeling

Mariia Zameshina (1, 2) , Olivier Teytaud (2) , Fabien Teytaud (3) , Vlad Hosu (4) , Nathanael Carraz (5) , Laurent Najman (1) , Markus Wagner (6)

1 LIGM - Laboratoire d'Informatique Gaspard-Monge
2 FAIR - Facebook AI Research [Paris]
3 LISIC - Laboratoire d'Informatique Signal et Image de la Côte d'Opale
4 Department of Computer and Information Science, University of Konstanz
5 Université d'Antananarivo
6 University of Adelaide

We design general-purpose algorithms for addressing fairness issues and mode collapse in generative modeling. More precisely, to design fair algorithms for as many sensitive variables as possible, including variables we might not be aware of, we assume no prior knowledge of sensitive variables: our algorithms use unsupervised fairness only, meaning no information related to the sensitive variables is used for our fairness-improving methods. All images of faces (even generated ones) have been removed to mitigate legal risks.

Mots clés: Generative modeling; neural networks; fairness

Domaines: Intelligence artificielle [cs.AI]

Mariia Zameshina, Olivier Teytaud, Fabien Teytaud, Vlad Hosu, Nathanael Carraz, et al.. Fairness in generative modeling. GECCO '22: Genetic and Evolutionary Computation Conference, Jul 2022, Boston Massachusetts, France. pp.320-323, ⟨10.1145/3520304.3528992⟩. ⟨hal-03793686⟩