Fairness-aware PageRank via Edge Reweighting

2025-12-08

AI Summary

This work addresses the lack of group fairness in PageRank, proposing a fairness-aware adaptation that preserves both the original network topology and restart vector. Methodologically, it formalizes *group-adaptivity fairness* based on group homophily and employs projected gradient descent to optimize edge weights of the transition probability matrix—enabling group-aware random walks that align with a target fair distribution. Crucially, the approach leverages a reweighting mechanism and non-convex optimization to achieve fairness through minimal, localized adjustments to edge weights, without structural modifications to the graph. Experiments across multiple benchmark datasets demonstrate that the method consistently outperforms state-of-the-art fair PageRank baselines in terms of inter-group ranking fairness, while maintaining computationally tractable overhead.

Technology Category

• Machine Learning: Learning Preferences or Rankings
• Game Theory and Economic Paradigms: Fair Division
• Search and Optimization: Distributed Search

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Link-analysis algorithms, such as PageRank, are instrumental in understanding the structural dynamics of networks by evaluating the importance of individual vertices based on their connectivity. Recently, with the rising importance of responsible AI, the question of fairness in link-analysis algorithms has gained traction. In this paper, we present a new approach for incorporating group fairness into the PageRank algorithm by reweighting the transition probabilities in the underlying transition matrix. We formulate the problem of achieving fair PageRank by seeking to minimize the fairness loss, which is the difference between the original group-wise PageRank distribution and a target PageRank distribution. We further define a group-adapted fairness notion, which accounts for group homophily by considering random walks with group-biased restart for each group. Since the fairness loss is non-convex, we propose an efficient projected gradient-descent method for computing locally-optimal edge weights. Unlike earlier approaches, we do not recommend adding new edges to the network, nor do we adjust the restart vector. Instead, we keep the topology of the underlying network unchanged and only modify the relative importance of existing edges. We empirically compare our approach with state-of-the-art baselines and demonstrate the efficacy of our method, where very small changes in the transition matrix lead to significant improvement in the fairness of the PageRank algorithm.

Problem

• Incorporating group fairness into PageRank via edge reweighting
• Minimizing fairness loss between original and target PageRank distributions
• Maintaining network topology while improving fairness through transition matrix adjustments

Innovation

• Reweighting transition probabilities for fairness
• Minimizing fairness loss with gradient descent
• Modifying edge weights without altering topology