Graph Posterior Network: Bayesian Predictive Uncertainty for Node Classification

Abstract

The interdependence between nodes in graphs is key to improve class predictions on nodes and utilized in approaches like Label Propagation (LP) or in Graph Neural Networks (GNN). Nonetheless, uncertainty estimation for non-independent node-level predictions is under-explored. In this work, we explore uncertainty quantification for node classification in three ways: (1) We derive three axioms explicitly characterizing the expected predictive uncertainty behavior in homophilic attributed graphs. (2) We propose a new model Graph Posterior Network (GPN) which explicitly performs Bayesian posterior updates for predictions on interdependent nodes. GPN provably obeys the proposed axioms. (3) We extensively evaluate GPN and a strong set of baselines on semi-supervised node classification including detection of anomalous features, and detection of left-out classes. GPN outperforms existing approaches for uncertainty estimation in the experiments.

Neurips 2021

References in corpus (12)

• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• Fast Graph Representation Learning with PyTorch Geometric
• FastGCN: Fast Learning with Graph Convolutional Networks via Importance Sampling
• Combining Label Propagation and Simple Models Out-performs Graph Neural Networks
• Unifying Graph Convolutional Neural Networks and Label Propagation
• BatchEnsemble: An Alternative Approach to Efficient Ensemble and Lifelong Learning
• Contrastive Training for Improved Out-of-Distribution Detection
• Uncertainty quantification of molecular property prediction with Bayesian neural networks
• Bayesian Graph Convolutional Neural Networks using Node Copying
• Bayesian Graph Convolutional Neural Networks Using Non-Parametric Graph Learning
• Uncertainty-aware Attention Graph Neural Network for Defending Adversarial Attacks
• I-GCN: Robust Graph Convolutional Network via Influence Mechanism