A Moral Framework for Understanding Fair ML Through Economic Models of Equality of Opportunity

Michele Loi, Hoda Heidari, Krishna P. Gummadi, Andreas Krause

Abstract:

This paper connects formal definitions of algorithmic fairness to economic and philosophical models of equality of opportunity. It shows how criteria such as predictive value parity and equality of odds can be interpreted as special cases of equality of opportunity, makes explicit the moral assumptions behind different fairness metrics, and proposes a family of fairness measures inspired by luck egalitarian theories.
Venue: FAT* ’19: Proceedings of the Conference on Fairness, Accountability, and Transparency, pp. 181-190