An Intersectional Definition of Fairness
This work addresses fairness for marginalized groups in AI systems by integrating intersectionality, offering a novel approach to a known bottleneck in fairness definitions.
The authors tackled the problem of defining fairness in ML/AI systems by incorporating intersectionality from Humanities, analyzing overlapping dimensions like gender, race, and disability. They developed a learning algorithm respecting these criteria, proving economic, privacy, and generalization guarantees, and demonstrated utility on census and COMPAS datasets.
We propose definitions of fairness in machine learning and artificial intelligence systems that are informed by the framework of intersectionality, a critical lens arising from the Humanities literature which analyzes how interlocking systems of power and oppression affect individuals along overlapping dimensions including gender, race, sexual orientation, class, and disability. We show that our criteria behave sensibly for any subset of the set of protected attributes, and we prove economic, privacy, and generalization guarantees. We provide a learning algorithm which respects our intersectional fairness criteria. Case studies on census data and the COMPAS criminal recidivism dataset demonstrate the utility of our methods.