When fairness metrics disagree operational implications for face recognition systems Exploring how to select the right fairness metrics for robust and reliable biometric evaluations

Get the scientific paper

You are using an obsolete browser (Internet Explorer < 11). For a safe user experience use the latest version.

When fairness metrics disagree operational implications for face recognition systems

Exploring how to select the right fairness metrics for robust and reliable biometric evaluations

When fairness metrics disagree operational implications for face recognition systems

What’s in the Scientific paper?

This scientific paper explores the challenges of assessing fairness in biometric systems and explains why different fairness metrics can lead to different conclusions.

It evaluates 19 fairness metrics across diverse real-world conditions and provides a practical framework to help organizations select the most appropriate metrics for robust and reliable fairness evaluations.


What you will learn

  • Why different fairness metrics can lead to different conclusions
  • How 19 fairness metrics perform under diverse real-world conditions
  • Why the right fairness metric depends on the type of bias being assessed
  • A practical framework for selecting appropriate fairness metrics for robust and reliable evaluations

Tell us a bit about yourself

*Required fields

You are using an obsolete browser (Internet Explorer < 11). For a safe user experience use the latest version.

Download

You might be interested in.

Explore the latest insights from the world of payments, smart mobility and open banking.
Share your challenge.

Our Fime experts are here to help you make innovation possible,
from defining, designing to delivering and testing your products
and services.

문의하기