Fairness of Classifiers in the Presence of Constraints between Features

TL;DR

This paper proposes a new fairness criterion for classifiers based on explanations that exclude protected features, considering feature constraints, and analyzes the complexity of testing fairness.

Contribution

It introduces a fairness definition based on explanations without protected features, accounting for feature constraints, and explores its theoretical properties and computational complexity.

Findings

Ignoring constraints can alter fairness assessments.

Three definitions of fairness are proposed and related.

Testing fairness under these definitions varies in computational complexity.

Abstract

In Machine Learning, an accepted definition of fairness of a decision taken by a classifier is that it should not depend on protected features, such as gender. Unfortunately, when constraints exist between features, such dependencies can be obscured by the constraints. To avoid this problem, we propose that a decision be considered fair if it has a fair explanation. We define a fair explanation as a prime-implicant reason for the decision that does not contain any protected feature (where the constraints are taken into account in the definition of prime-implicant). Surprisingly, ignoring constraints can completely change the fairness of a decision (according to this definition) even in the absence of constraints between protected and unprotected features. Three possible definitions of fairness of a classifier are that for all its decisions (1) there are only fair explanations, (2) there is at least one fair explanation, or (3) changing protected features does not change the outcome. We identify the relationships between these different definitions of fairness and study the computational complexity of testing fairness of classifiers.