Pairwise Fairness for Ordinal Regression
This work addresses fairness for ordinal regression, which is an incremental extension of existing fairness concepts to a new problem domain.
The paper tackles fairness in ordinal regression by adapting two fairness notions from fair ranking and proposing a training strategy for an approximately fair threshold model predictor, demonstrating effectiveness in extensive experiments.
We initiate the study of fairness for ordinal regression. We adapt two fairness notions previously considered in fair ranking and propose a strategy for training a predictor that is approximately fair according to either notion. Our predictor has the form of a threshold model, composed of a scoring function and a set of thresholds, and our strategy is based on a reduction to fair binary classification for learning the scoring function and local search for choosing the thresholds. We provide generalization guarantees on the error and fairness violation of our predictor, and we illustrate the effectiveness of our approach in extensive experiments.