Fair Interpretable Learning via Correction Vectors

TL;DR

This paper introduces a transparent method for fair representation learning using correction vectors, enabling interpretability without sacrificing model performance.

Contribution

It proposes a novel framework that employs correction vectors for fair learning, enhancing interpretability compared to traditional neural network debiasing methods.

Findings

Fairness does not compromise performance in the proposed framework.

Correction vectors provide explicit feature-level adjustments.

Method enhances interpretability of fair representation learning.

Abstract

Neural network architectures have been extensively employed in the fair representation learning setting, where the objective is to learn a new representation for a given vector which is independent of sensitive information. Various "representation debiasing" techniques have been proposed in the literature. However, as neural networks are inherently opaque, these methods are hard to comprehend, which limits their usefulness. We propose a new framework for fair representation learning which is centered around the learning of "correction vectors", which have the same dimensionality as the given data vectors. The corrections are then simply summed up to the original features, and can therefore be analyzed as an explicit penalty or bonus to each feature. We show experimentally that a fair representation learning problem constrained in such a way does not impact performance.