False-Positive Rate (FMR)
Definition
In biometric verification: the probability that a comparison of images of two different individuals is incorrectly declared a match. In forensic identification, an elevated FMR for a demographic group creates a disproportionate risk that innocent members of that group will be presented as candidate matches to investigators.
- Context
- Biometric verification and demographic bias assessment
- Formula
- Probability a different-person comparison is called a match
- Bias concern
- FMR can vary by demographic group at a fixed threshold
- Consequence
- Innocent people wrongly surfaced as candidate matches
Common questions
Why does an FMR gap between demographic groups matter operationally?+
If a system's FMR is higher for one demographic at the threshold in use, members of that group face a higher chance of being wrongly flagged as a candidate match than others, even though the system's overall reported accuracy looks acceptable.
Can demographic FMR differences be fixed by simply lowering the threshold?+
A single global threshold change reduces false matches for everyone but does not close a gap between groups, and it raises false non-matches across the board; addressing a demographic disparity typically requires evaluating performance per subgroup, not just adjusting one overall setting.
Related terms
- AFIS (Automated Fingerprint Identification System)
- An automated system that compares a fingerprint query (typically a latent print from a crime scene) against a reference database to identify...
- Clearview AI
- New York company that built a face-recognition database of approximately 30 billion images scraped from public websites without consent; subject to BIPA...
- Daubert Standard
- The US federal evidentiary standard (Daubert v. Merrell Dow Pharmaceuticals, 1993) requiring that expert testimony be based on scientifically valid methods with...
- Disparate Impact
- A legal and statistical concept describing a policy or practice that has a disproportionate adverse effect on a protected group, regardless of...
- EU AI Act Art 10 (Training Data Requirements)
- Provision requiring that training, validation, and testing datasets for high-risk AI systems be relevant, representative, free of errors, and complete, with appropriate...
- Fitzpatrick Scale
- A skin-tone classification system originally developed by dermatologist Thomas B. Fitzpatrick in 1975 (initially covering four types for UV response in phototherapy),...
- Gender Shades (2018)
- Study by Joy Buolamwini and Timnit Gebru published at FAccT 2018 that measured accuracy disparities in three commercial face-recognition APIs across intersectional...
- NIST FRVT (Face Recognition Vendor Test)
- Ongoing NIST evaluation of commercial face-recognition algorithms, with the 2019 Demographic Effects report covering 189 algorithms from 99 developers; confirms Gender Shades...
- Robert Williams Case (2020)
- First publicly documented wrongful arrest in the US attributed to a face-recognition false positive: Detroit police arrested Robert Williams, a Black man,...