English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The development of face recognition algorithms by academic and commercialorganizations is growing rapidly due to the onset of deep learning and thewidespread availability of training data. Though tests of face recognitionalgorithm performance indicate yearly performance gains, error rates for manyof these systems differ based on the demographic composition of the test set.These "demographic differentials" in algorithm performance can contribute tounequal or unfair outcomes for certain groups of people, raising concerns withincreased worldwide adoption of face recognition systems. Consequently,regulatory bodies in both the United States and Europe have proposed new rulesrequiring audits of biometric systems for "discriminatory impacts" (EuropeanUnion Artificial Intelligence Act) and "fairness" (U.S. Federal TradeCommission). However, no standard for measuring fairness in biometric systemsyet exists. This paper characterizes two proposed measures of face recognitionalgorithm fairness (fairness measures) from scientists in the U.S. and Europe.We find that both proposed methods are challenging to interpret when applied todisaggregated face recognition error rates as they are commonly experienced inpractice. To address this, we propose a set of interpretability criteria,termed the Functional Fairness Measure Criteria (FFMC), that outlines a set ofproperties desirable in a face recognition algorithm fairness measure. Wefurther develop a new fairness measure, the Gini Aggregation Rate for BiometricEquitability (GARBE), and show how, in conjunction with the Paretooptimization, this measure can be used to select among alternative algorithmsbased on the accuracy/fairness trade-space. Finally, we have open-sourced ourdataset of machine-readable, demographically disaggregated error rates. Webelieve this is currently the largest open-source dataset of its kind.

Evaluating Proposed Fairness Models for Face Recognition Algorithms