FairSample: Training Fair and Accurate Graph Convolutional Neural Networks Efficiently

Abstract

Fairness in Graph Convolutional Neural Networks (GCNs) becomes a more and more important concern as GCNs are adopted in many crucial applications. Societal biases against sensitive groups may exist in many real world graphs. GCNs trained on those graphs may be vulnerable to being affected by such biases. In this paper, we adopt the well-known fairness notion of demographic parity and tackle the challenge of training fair and accurate GCNs efficiently. We present an in-depth analysis on how graph structure bias, node attribute bias, and model parameters may affect the demographic parity of GCNs. Our insights lead to FairSample, a framework that jointly mitigates the three types of biases. We employ two intuitive strategies to rectify graph structures. First, we inject edges across nodes that are in different sensitive groups but similar in node features. Second, to enhance model fairness and retain model quality, we develop a learnable neighbor sampling policy using reinforcement learning. To address the bias in node features and model parameters, FairSample is complemented by a regularization objective to optimize fairness.

Accepted by TKDE 2023

References in corpus (25)

• Spectral Normalization for Generative Adversarial Networks
• An Introduction to Deep Reinforcement Learning
• Simplifying Graph Convolutional Networks
• Graph Contrastive Learning with Adaptive Augmentation
• Counterfactual Fairness
• Disentangled Graph Collaborative Filtering
• Towards Deeper Graph Neural Networks
• LightGCN: Simplifying and Powering Graph Convolution Network for Recommendation
• Simple and Deep Graph Convolutional Networks
• Scaling Graph Neural Networks with Approximate PageRank
• EDITS: Modeling and Mitigating Data Bias for Graph Neural Networks
• Graph Neural Networks in Recommender Systems: A Survey
• Investigating and Mitigating Degree-Related Biases in Graph Convolutional Networks
• Compositional Fairness Constraints for Graph Embeddings
• Learning Fair Node Representations with Graph Counterfactual Fairness
• RawlsGCN: Towards Rawlsian Difference Principle on Graph Convolutional Network
• Putting Fairness Principles into Practice: Challenges, Metrics, and Improvements
• Subgroup Generalization and Fairness of Graph Neural Networks
• Bandit Samplers for Training Graph Neural Networks
• Enhancing the Accuracy and Fairness of Human Decision Making
• Training Fair Models in Federated Learning without Data Privacy Infringement
• NRGNN: Learning a Label Noise-Resistant Graph Neural Network on Sparsely and Noisily Labeled Graphs
• Fair Node Representation Learning via Adaptive Data Augmentation
• All of the Fairness for Edge Prediction with Optimal Transport
• Fair Data Representation for Machine Learning at the Pareto Frontier