Operational Fairness for Facial Authentication Systems - Département Sciences économiques et sociales Access content directly
Special Issue Year : 2022

Operational Fairness for Facial Authentication Systems

Abstract

How to design a facial authentication system taking into account both performance and fairness? We consider the choices that a developer makes when coding such a system, such as the training parameters, the architecture of the neural network or the authentication threshold. We evaluate their impact on the global fairness of the system showing that fairness is not only affected by the training data but also by the multiple choices that are made when coding the model.
Fichier principal
Vignette du fichier
dtis22200.pdf (371.25 Ko) Télécharger le fichier
Origin : Publisher files allowed on an open archive

Dates and versions

hal-03837716 , version 1 (14-12-2022)

Identifiers

  • HAL Id : hal-03837716 , version 1

Cite

Mélanie Gornet, Claude Kirchner, Catherine Tessier. Operational Fairness for Facial Authentication Systems. 2022. ⟨hal-03837716⟩
112 View
51 Download

Share

Gmail Facebook X LinkedIn More