Bipartite Ranking Fairness through a Model Agnostic Ordering Adjustment

TL;DR

This paper introduces xOrder, a model-agnostic post-processing method for bipartite ranking that enhances fairness across protected groups while preserving ranking performance, applicable to multiple fairness metrics and group types.

Contribution

The paper presents a novel, flexible post-processing framework for fairness in bipartite ranking that optimizes a utility-weighted warping path via dynamic programming, compatible with various models and metrics.

Findings

xOrder improves fairness-utility balance across datasets

It reduces score distribution shifts between groups

Maintains robustness with fewer samples and distribution differences

Abstract

Algorithmic fairness has been a serious concern and received lots of interest in machine learning community. In this paper, we focus on the bipartite ranking scenario, where the instances come from either the positive or negative class and the goal is to learn a ranking function that ranks positive instances higher than negative ones. While there could be a trade-off between fairness and performance, we propose a model agnostic post-processing framework xOrder for achieving fairness in bipartite ranking and maintaining the algorithm classification performance. In particular, we optimize a weighted sum of the utility as identifying an optimal warping path across different protected groups and solve it through a dynamic programming process. xOrder is compatible with various classification models and ranking fairness metrics, including supervised and unsupervised fairness metrics. In addition to binary groups, xOrder can be applied to multiple protected groups. We evaluate our proposed algorithm on four benchmark data sets and two real-world patient electronic health record repositories. xOrder consistently achieves a better balance between the algorithm utility and ranking fairness on a variety of datasets with different metrics. From the visualization of the calibrated ranking scores, xOrder mitigates the score distribution shifts of different groups compared with baselines. Moreover, additional analytical results verify that xOrder achieves a robust performance when faced with fewer samples and a bigger difference between training and testing ranking score distributions.