Disentangling, Amplifying, and Debiasing: Learning Disentangled Representations for Fair Graph Neural Networks

TL;DR

This paper introduces DAB-GNN, a novel framework for fair graph neural networks that disentangles, amplifies, and debiases biases in node attributes and graph structure, improving fairness without sacrificing accuracy.

Contribution

The paper presents a new GNN framework that explicitly disentangles and amplifies biases before debiasing, which enhances fairness and maintains high accuracy in graph learning tasks.

Findings

DAB-GNN outperforms ten state-of-the-art methods in fairness and accuracy.

Disentanglement and amplification improve bias mitigation effectiveness.

Extensive experiments validate the framework's robustness across datasets.

Abstract

Graph Neural Networks (GNNs) have become essential tools for graph representation learning in various domains, such as social media and healthcare. However, they often suffer from fairness issues due to inherent biases in node attributes and graph structure, leading to unfair predictions. To address these challenges, we propose a novel GNN framework, DAB-GNN, that Disentangles, Amplifies, and deBiases attribute, structure, and potential biases in the GNN mechanism. DAB-GNN employs a disentanglement and amplification module that isolates and amplifies each type of bias through specialized disentanglers, followed by a debiasing module that minimizes the distance between subgroup distributions. Extensive experiments on five datasets demonstrate that DAB-GNN significantly outperforms ten state-of-the-art competitors in terms of achieving an optimal balance between accuracy and fairness. The codebase of DAB-GNN is available at https://github.com/Bigdasgit/DAB-GNN