Projection to Fairness in Statistical Learning
This work addresses fairness in statistical learning for regression tasks, offering a method to balance fairness and accuracy, but it is incremental as it builds on existing optimal transport techniques.
The paper tackles the problem of making regression estimators fair while minimizing loss in prediction accuracy by defining a 'projection to fairness' as the closest fair estimator, using optimal transport for efficient post-processing and quantifying the fairness cost in accuracy terms.
In the context of regression, we consider the fundamental question of making an estimator fair while preserving its prediction accuracy as much as possible. To that end, we define its projection to fairness as its closest fair estimator in a sense that reflects prediction accuracy. Our methodology leverages tools from optimal transport to construct efficiently the projection to fairness of any given estimator as a simple post-processing step. Moreover, our approach precisely quantifies the cost of fairness, measured in terms of prediction accuracy.