Improving Fairness in Graph Neural Networks via Mitigating Sensitive Attribute Leakage

Abstract

Graph Neural Networks (GNNs) have shown great power in learning node representations on graphs. However, they may inherit historical prejudices from training data, leading to discriminatory bias in predictions. Although some work has developed fair GNNs, most of them directly borrow fair representation learning techniques from non-graph domains without considering the potential problem of sensitive attribute leakage caused by feature propagation in GNNs. However, we empirically observe that feature propagation could vary the correlation of previously innocuous non-sensitive features to the sensitive ones. This can be viewed as a leakage of sensitive information which could further exacerbate discrimination in predictions. Thus, we design two feature masking strategies according to feature correlations to highlight the importance of considering feature propagation and correlation variation in alleviating discrimination. Motivated by our analysis, we propose Fair View Graph Neural Network (FairVGNN) to generate fair views of features by automatically identifying and masking sensitive-correlated features considering correlation variation after feature propagation. Given the learned fair views, we adaptively clamp weights of the encoder to avoid using sensitive-related features. Experiments on real-world datasets demonstrate that FairVGNN enjoys a better trade-off between model utility and fairness. Our code is publicly available at https://github.com/YuWVandy/FairVGNN.

References in corpus (8)

• Towards Deeper Graph Neural Networks
• Simple and Deep Graph Convolutional Networks
• Self-supervised Learning on Graphs: Deep Insights and New Direction
• Compositional Fairness Constraints for Graph Embeddings
• Investigating and Mitigating Degree-Related Biases in Graph Convolutional Networks
• Feature Overcorrelation in Deep Graph Neural Networks: A New Perspective
• Fairness-Aware Node Representation Learning
• Distance-wise Prototypical Graph Neural Network in Node Imbalance Classification

Cited by in corpus (13)

• A Comprehensive Survey on Trustworthy Graph Neural Networks: Privacy, Robustness, Fairness, and Explainability
• Condensing Graphs via One-Step Gradient Matching
• Collaboration-Aware Graph Convolutional Network for Recommender Systems
• Feature Overcorrelation in Deep Graph Neural Networks: A New Perspective
• Towards Mitigating Dimensional Collapse of Representations in Collaborative Filtering
• Masked Graph Transformer for Large-Scale Recommendation
• Endowing Pre-trained Graph Models with Provable Fairness
• TinyKG: Memory-Efficient Training Framework for Knowledge Graph Neural Recommender Systems
• Adversarial Collaborative Filtering for Free
• MAPPING: Debiasing Graph Neural Networks for Fair Node Classification with Limited Sensitive Information Leakage
• Enhancing Transformers without Self-supervised Learning: A Loss Landscape Perspective in Sequential Recommendation
• Fairness without Demographics through Learning Graph of Gradients
• Hessian-aware Quantized Node Embeddings for Recommendation