Pairwise Fairness for Ordinal Regression

TL;DR

This paper introduces a novel approach to fairness in ordinal regression, adapting fairness notions from ranking, and proposes a threshold-based predictor with theoretical guarantees and experimental validation.

Contribution

It develops a new fairness-aware ordinal regression method using a reduction to fair binary classification and local search for thresholds, with formal guarantees.

Findings

Effective in achieving fairness in ordinal regression tasks

Provides theoretical generalization guarantees

Demonstrates strong empirical performance

Abstract

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.