BeMap: Balanced Message Passing for Fair Graph Neural Network
This addresses fairness issues in graph neural networks for applications like social network analysis, though it is incremental as it builds on existing fairness methods by focusing on message passing.
The paper tackles bias amplification in graph neural networks by analyzing how unbalanced neighbor distributions across demographic groups during message passing increase unfairness, and proposes BeMap, a balance-aware sampling method that reduces bias by up to 30% while preserving classification accuracy.
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.