The key highlights and insights of this content are:
The authors propose a new interpretable measure of unfairness that quantifies the fraction of the population treated differently from an underlying baseline treatment. This measure has a clear probabilistic interpretation and satisfies desirable properties.
The authors generalize the equalized odds fairness criterion, originally defined for binary classification, to the multiclass setting. The multiclass equalized odds criterion requires that the confusion matrices of the classifier be the same across all sub-populations.
When the confusion matrices of the classifier in each sub-population are known, the authors show how to calculate the unfairness measure exactly for binary classification, and provide algorithms to obtain upper and lower bounds for the multiclass case.
In the more challenging case where only label proportions are available, without access to the confusion matrices, the authors propose methods to lower-bound the unfairness and the error of the classifier. This is done by exploiting the inherent trade-off between accuracy and fairness.
The authors report experiments demonstrating the effectiveness of their proposed methods in various scenarios, including applications such as election prediction, disease diagnosis, and analysis of access to healthcare.
toiselle kielelle
lähdeaineistosta
arxiv.org
Tärkeimmät oivallukset
by Sivan Sabato... klo arxiv.org 04-09-2024
https://arxiv.org/pdf/2206.03234.pdfSyvällisempiä Kysymyksiä