Learning Fair Graph Representations with Multi-view Information Bottleneck
This addresses fairness issues in graph neural networks for applications like social networks or recommendation systems, offering a novel method to handle multiple bias sources, though it is incremental in improving existing fairness techniques.
The paper tackled the problem of graph neural networks amplifying biases in training data by proposing FairMIB, a multi-view information bottleneck framework that decomposes graphs into feature, structural, and diffusion views to mitigate biases, achieving state-of-the-art performance on utility and fairness metrics across five benchmark datasets.
Graph neural networks (GNNs) excel on relational data by passing messages over node features and structure, but they can amplify training data biases, propagating discriminatory attributes and structural imbalances into unfair outcomes. Many fairness methods treat bias as a single source, ignoring distinct attribute and structure effects and leading to suboptimal fairness and utility trade-offs. To overcome this challenge, we propose FairMIB, a multi-view information bottleneck framework designed to decompose graphs into feature, structural, and diffusion views for mitigating complexity biases in GNNs. Especially, the proposed FairMIB employs contrastive learning to maximize cross-view mutual information for bias-free representation learning. It further integrates multi-perspective conditional information bottleneck objectives to balance task utility and fairness by minimizing mutual information with sensitive attributes. Additionally, FairMIB introduces an inverse probability-weighted (IPW) adjacency correction in the diffusion view, which reduces the spread of bias propagation during message passing. Experiments on five real-world benchmark datasets demonstrate that FairMIB achieves state-of-the-art performance across both utility and fairness metrics.