Fairness-Aware Graph Representation Learning with Limited Demographic Information

2025-11-17

AI Summary

Graph neural networks (GNNs) often suffer from unfairness when sensitive attributes (e.g., race, gender) are partially missing—a common yet underexplored scenario in fair representation learning. Method: We propose a proxy-driven fair graph representation learning framework. First, it constructs reliable proxy variables from available demographic information. Second, it introduces a cross-group consistency embedding mechanism coupled with an adaptive confidence-weighted strategy to dynamically balance fairness and predictive utility. Contribution/Results: Theoretically, we establish the first bias upper bound guarantee for fair graph learning under limited sensitive information. Empirically, on multiple benchmark datasets, our framework significantly reduces inter-group representation shift (average reduction: 32.7%) and prediction bias (average reduction: 28.4%), while maintaining state-of-the-art predictive performance. These results demonstrate its effectiveness, robustness, and generalizability.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Reasoning under Uncertainty: Graphical Models
• Computer Vision: Bias, Fairness & Privacy

Application Category

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

Abstract

Ensuring fairness in Graph Neural Networks is fundamental to promoting trustworthy and socially responsible machine learning systems. In response, numerous fair graph learning methods have been proposed in recent years. However, most of them assume full access to demographic information, a requirement rarely met in practice due to privacy, legal, or regulatory restrictions. To this end, this paper introduces a novel fair graph learning framework that mitigates bias in graph learning under limited demographic information. Specifically, we propose a mechanism guided by partial demographic data to generate proxies for demographic information and design a strategy that enforces consistent node embeddings across demographic groups. In addition, we develop an adaptive confidence strategy that dynamically adjusts each node's contribution to fairness and utility based on prediction confidence. We further provide theoretical analysis demonstrating that our framework, FairGLite, achieves provable upper bounds on group fairness metrics, offering formal guarantees for bias mitigation. Through extensive experiments on multiple datasets and fair graph learning frameworks, we demonstrate the framework's effectiveness in both mitigating bias and maintaining model utility.

Problem

• Mitigating bias in graph learning with limited demographic data access
• Generating demographic proxies and ensuring consistent node embeddings
• Achieving group fairness guarantees while maintaining model utility

Innovation

• Generates demographic proxies using partial data
• Enforces consistent embeddings across demographic groups
• Dynamically adjusts fairness contributions via confidence strategy