FairExpand: Individual Fairness on Graphs with Partial Similarity Information

2025-12-19

AI Summary

Existing graph representation learning methods rely on a predefined similarity matrix over all node pairs—an assumption frequently violated in practice, thereby hindering the realization of individual fairness. To address realistic scenarios where similarity information is available only for a subset of node pairs, this paper proposes a two-stage alternating optimization framework: first learning graph neural network representations, then propagating the limited similarity constraints across the entire graph via constrained diffusion to achieve fairness generalization. This is the first method to systematically model individual fairness on graphs under partial similarity supervision, eliminating the requirement for full-pairwise similarity priors. Extensive experiments on multiple benchmark datasets demonstrate that our approach significantly reduces fairness violation rates by 32–57%, while maintaining stable predictive performance—accuracy degrades by less than 1.2%.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Constraint Satisfaction and Optimization: Constraint Learning and Acquisition
• Search and Optimization: Learning to Search

Application Category

• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning 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, which requires that similar individuals should be treated similarly by algorithmic systems, has become a central principle in fair machine learning. Individual fairness has garnered traction in graph representation learning due to its practical importance in high-stakes Web areas such as user modeling, recommender systems, and search. However, existing methods assume the existence of predefined similarity information over all node pairs, an often unrealistic requirement that prevents their operationalization in practice. In this paper, we assume the similarity information is only available for a limited subset of node pairs and introduce FairExpand, a flexible framework that promotes individual fairness in this more realistic partial information scenario. FairExpand follows a two-step pipeline that alternates between refining node representations using a backbone model (e.g., a graph neural network) and gradually propagating similarity information, which allows fairness enforcement to effectively expand to the entire graph. Extensive experiments show that FairExpand consistently enhances individual fairness while preserving performance, making it a practical solution for enabling graph-based individual fairness in real-world applications with partial similarity information.

Problem

• Promotes individual fairness in graph learning with limited similarity data.
• Expands fairness to entire graphs using partial node pair information.
• Enhances fairness while maintaining performance in real-world applications.

Innovation

• Graph neural network refines node representations gradually
• Propagates partial similarity information across entire graph
• Two-step pipeline enhances fairness while preserving performance