BeMap: Balanced Message Passing for Fair Graph Neural Network

TL;DR

This paper introduces BeMap, a fair message passing method for graph neural networks that reduces bias amplification caused by unbalanced demographic groups in neighbors, improving fairness without sacrificing accuracy.

Contribution

We propose BeMap, a novel balanced message passing approach that explicitly addresses bias amplification in GNNs through a balance-aware sampling strategy.

Findings

BeMap effectively reduces bias in node classification tasks.

BeMap maintains high classification accuracy while improving fairness.

Theoretical analysis supports the bias mitigation capability of BeMap.

Abstract

Fairness in graph neural networks has been actively studied recently. However, existing works often do not explicitly consider the role of message passing in introducing or amplifying the bias. In this paper, we first investigate the problem of bias amplification in message passing. We empirically and theoretically demonstrate that message passing could amplify the bias when the 1-hop neighbors from different demographic groups are unbalanced. Guided by such analyses, we propose BeMap, a fair message passing method, that leverages a balance-aware sampling strategy to balance the number of the 1-hop neighbors of each node among different demographic groups. Extensive experiments on node classification demonstrate the efficacy of BeMap in mitigating bias while maintaining classification accuracy. The code is available at https://github.com/xiaolin-cs/BeMap.