On Generalized Degree Fairness in Graph Neural Networks

TL;DR

This paper introduces Deg-FairGNN, a novel GNN framework that addresses degree bias caused by neighborhood structure diversity, improving fairness and accuracy in node classification tasks.

Contribution

It defines and generalizes degree bias in GNNs and proposes a learnable debiasing method to mitigate this bias during neighborhood aggregation.

Findings

Effective in reducing degree bias across datasets

Improves fairness metrics without sacrificing accuracy

Demonstrates robustness on multiple benchmarks

Abstract

Conventional graph neural networks (GNNs) are often confronted with fairness issues that may stem from their input, including node attributes and neighbors surrounding a node. While several recent approaches have been proposed to eliminate the bias rooted in sensitive attributes, they ignore the other key input of GNNs, namely the neighbors of a node, which can introduce bias since GNNs hinge on neighborhood structures to generate node representations. In particular, the varying neighborhood structures across nodes, manifesting themselves in drastically different node degrees, give rise to the diverse behaviors of nodes and biased outcomes. In this paper, we first define and generalize the degree bias using a generalized definition of node degree as a manifestation and quantification of different multi-hop structures around different nodes. To address the bias in the context of node classification, we propose a novel GNN framework called Generalized Degree Fairness-centric Graph Neural Network (Deg-FairGNN). Specifically, in each GNN layer, we employ a learnable debiasing function to generate debiasing contexts, which modulate the layer-wise neighborhood aggregation to eliminate the degree bias originating from the diverse degrees among nodes. Extensive experiments on three benchmark datasets demonstrate the effectiveness of our model on both accuracy and fairness metrics.