SaGIF: Improving Individual Fairness in Graph Neural Networks via Similarity Encoding

2025-06-23

AI Summary

Existing graph neural networks (GNNs) lack individual fairness—i.e., similar nodes should receive similar predictions—due to two key bottlenecks: unclear sources of unfairness and incomplete modeling of node similarity. Method: We propose a similarity-encoding-based fairness-enhancement framework. Its core innovation is the formal definition of a “similarity consistency” principle, coupled with a dual-path similarity metric that jointly encodes topological structure and node features, enabling learning of fairness-aware, node-wise representations. Crucially, fairness constraints are explicitly decoupled from the prediction task while preserving model utility. Results: Extensive experiments on multiple real-world graph datasets demonstrate that our method significantly outperforms state-of-the-art fair GNNs, achieving simultaneous improvements in individual fairness and predictive performance stability.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Philosophy and Ethics of AI: Bias, Fairness & Equity

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Social Networks and Social Media: Fairness and bias in social network and social media analysis

Abstract

Individual fairness (IF) in graph neural networks (GNNs), which emphasizes the need for similar individuals should receive similar outcomes from GNNs, has been a critical issue. Despite its importance, research in this area has been largely unexplored in terms of (1) a clear understanding of what induces individual unfairness in GNNs and (2) a comprehensive consideration of identifying similar individuals. To bridge these gaps, we conduct a preliminary analysis to explore the underlying reason for individual unfairness and observe correlations between IF and similarity consistency, a concept introduced to evaluate the discrepancy in identifying similar individuals based on graph structure versus node features. Inspired by our observations, we introduce two metrics to assess individual similarity from two distinct perspectives: topology fusion and feature fusion. Building upon these metrics, we propose Similarity-aware GNNs for Individual Fairness, named SaGIF. The key insight behind SaGIF is the integration of individual similarities by independently learning similarity representations, leading to an improvement of IF in GNNs. Our experiments on several real-world datasets validate the effectiveness of our proposed metrics and SaGIF. Specifically, SaGIF consistently outperforms state-of-the-art IF methods while maintaining utility performance. Code is available at: https://github.com/ZzoomD/SaGIF.

Problem

• Understanding causes of individual unfairness in GNNs
• Identifying similar individuals comprehensively in GNNs
• Improving individual fairness via similarity-aware representations

Innovation

• Similarity encoding for individual fairness
• Topology and feature fusion metrics
• Independent similarity representation learning