Multi-Output Distributional Fairness via Post-Processing

TL;DR

This paper introduces a post-processing method using optimal transport to improve distributional fairness in multi-output machine learning models, addressing limitations of existing single-output focused approaches.

Contribution

It extends distributional fairness post-processing to multi-output models using Wasserstein barycenters and kernel regression, enabling broader applicability.

Findings

Effective in multi-task and multi-class classification tasks

Outperforms baseline methods in fairness metrics

Scalable to out-of-sample data

Abstract

The post-processing approaches are becoming prominent techniques to enhance machine learning models' fairness because of their intuitiveness, low computational cost, and excellent scalability. However, most existing post-processing methods are designed for task-specific fairness measures and are limited to single-output models. In this paper, we introduce a post-processing method for multi-output models, such as the ones used for multi-task/multi-class classification and representation learning, to enhance a model's distributional parity, a task-agnostic fairness measure. Existing methods for achieving distributional parity rely on the (inverse) cumulative density function of a model's output, restricting their applicability to single-output models. Extending previous works, we propose to employ optimal transport mappings to move a model's outputs across different groups towards their empirical Wasserstein barycenter. An approximation technique is applied to reduce the complexity of computing the exact barycenter and a kernel regression method is proposed to extend this process to out-of-sample data. Our empirical studies evaluate the proposed approach against various baselines on multi-task/multi-class classification and representation learning tasks, demonstrating the effectiveness of the proposed approach.