FairGC: Fairness-aware Graph Condensation

2026-03-30

AI Summary

Existing graph compression methods often overlook fairness, thereby exacerbating group biases in sensitive applications such as credit scoring. This work proposes FairGC, a novel framework that explicitly integrates fairness into the graph distillation process. FairGC jointly optimizes structural fidelity and fairness through distribution-preserving compression, spectral encoding based on Laplacian eigendecomposition, and a fairness-enhancing neural architecture that fuses multi-domain information with label-smoothed curriculum learning. The method simultaneously preserves the joint distribution of labels and sensitive attributes while leveraging spectral structures to improve fair predictive performance. Evaluated on four real-world datasets, FairGC significantly reduces disparities in statistical parity and equal opportunity while maintaining high accuracy, outperforming existing approaches.

Abstract

Graph condensation (GC) has become a vital strategy for scaling Graph Neural Networks by compressing massive datasets into small, synthetic node sets. While current GC methods effectively maintain predictive accuracy, they are primarily designed for utility and often ignore fairness constraints. Because these techniques are bias-blind, they frequently capture and even amplify demographic disparities found in the original data. This leads to synthetic proxies that are unsuitable for sensitive applications like credit scoring or social recommendations. To solve this problem, we introduce FairGC, a unified framework that embeds fairness directly into the graph distillation process. Our approach consists of three key components. First, a Distribution-Preserving Condensation module synchronizes the joint distributions of labels and sensitive attributes to stop bias from spreading. Second, a Spectral Encoding module uses Laplacian eigen-decomposition to preserve essential global structural patterns. Finally, a Fairness-Enhanced Neural Architecture employs multi-domain fusion and a label-smoothing curriculum to produce equitable predictions. Rigorous evaluations on four real-world datasets, show that FairGC provides a superior balance between accuracy and fairness. Our results confirm that FairGC significantly reduces disparity in Statistical Parity and Equal Opportunity compared to existing state-of-the-art condensation models. The codes are available at https://github.com/LuoRenqiang/FairGC.

Problem

• Graph Condensation
• Fairness
• Bias Amplification
• Synthetic Data
• Demographic Disparity

Innovation

• Fairness-aware Graph Condensation
• Distribution-Preserving Condensation
• Spectral Encoding
• Multi-domain Fusion
• Label-smoothing Curriculum