People are Not Coins: A Reply to Hedden.
17 Pages Posted: 3 Jun 2021 Last revised: 23 Jun 2021
Date Written: June 1, 2021
This paper is a reply to "On Statistical Criteria of Algorithmic Fairness," by Brian Hedden. We question the significance of arguing that many group fairness criteria discussed in the machine learning literature are not necessary conditions for the fairness of predictions or decisions based on them. We show that it may be true, in general, that F is not a necessary condition for the fairness of all predictions (or decisions based on them). And yet, compatibly with this, most predictions or decisions involving people could be unfair if they violate the statistical fairness constraint F.
Keywords: algorithms, fairness, justice, statistics
Suggested Citation: Suggested Citation