Fair and Interpretable Decision Rules for Binary Classification

Citation:

Lawless C, Günlük O. Fair and Interpretable Decision Rules for Binary Classification, in IJCAI 2021 Workshop on AI for Social Good. ; 2021.

Abstract:

In this paper we consider the problem of building Boolean rule sets in disjunctive normal form (DNF), an interpretable model for binary classification, subject to fairness constraints. We formulate the problem as an integer program that maximizes classification accuracy with explicit constraints on two different measures of classification parity: equality of opportunity, and equalized odds. A column generation framework, with a novel formulation, is used to efficiently search over exponentially many possible rules, eliminating the need for heuristic rule mining. Compared to CART and Logistic Regression, two interpretable machine learning algorithms, our method produces interpretable classifiers that have superior performance with respect to both fair- ness metrics.