<ul><li>This paper discusses the issue of fair probabilistic binary classification with binary protected groups.</li><li>The objective is to design a fair classifier that is fair to both protected groups, irrespective of the threshold used by the practitioner.</li><li>The proposed method, called FROC, introduces a threshold query model on ROC curves to transform a potentially unfair classifier's output to a fair classifier.</li><li>The algorithm achieves the theoretical optimal guarantees and is evaluated on various real-world datasets.</li></ul>

FROC: Building Fair ROC from a Trained Classifier

Discover more