An Empirical Study on Learning Fairness Metrics for COMPAS Data with Human Supervision
This provides an empirical tool for understanding human supervision in fairness metrics, but it is incremental as it builds on prior metric learning methods without directly solving the fairness metric learning problem.
The paper tackles the challenge of specifying similarity metrics for individual fairness in criminal recidivism prediction by learning metrics from human-annotated COMPAS data, showing that learned metrics outperform Euclidean and Precision metrics under various criteria.
The notion of individual fairness requires that similar people receive similar treatment. However, this is hard to achieve in practice since it is difficult to specify the appropriate similarity metric. In this work, we attempt to learn such similarity metric from human annotated data. We gather a new dataset of human judgments on a criminal recidivism prediction (COMPAS) task. By assuming the human supervision obeys the principle of individual fairness, we leverage prior work on metric learning, evaluate the performance of several metric learning methods on our dataset, and show that the learned metrics outperform the Euclidean and Precision metric under various criteria. We do not provide a way to directly learn a similarity metric satisfying the individual fairness, but to provide an empirical study on how to derive the similarity metric from human supervisors, then future work can use this as a tool to understand human supervision.