Predict then Propagate: Graph Neural Networks meet Personalized PageRank

Abstract

Neural message passing algorithms for semi-supervised classification on graphs have recently achieved great success. However, for classifying a node these methods only consider nodes that are a few propagation steps away and the size of this utilized neighborhood is hard to extend. In this paper, we use the relationship between graph convolutional networks (GCN) and PageRank to derive an improved propagation scheme based on personalized PageRank. We utilize this propagation procedure to construct a simple model, personalized propagation of neural predictions (PPNP), and its fast approximation, APPNP. Our model's training time is on par or faster and its number of parameters on par or lower than previous models. It leverages a large, adjustable neighborhood for classification and can be easily combined with any neural network. We show that this model outperforms several recently proposed methods for semi-supervised classification in the most thorough study done so far for GCN-like models. Our implementation is available online.

Published as a conference paper at ICLR 2019. Author name changed from Johannes Klicpera to Johannes Gasteiger

Cited by in corpus (62)

• Graph Neural Networks with convolutional ARMA filters
• Scaling Graph Neural Networks with Approximate PageRank
• A Comprehensive Survey on Deep Graph Representation Learning
• Attention-based graph neural networks: a survey
• Graph Attention Multi-Layer Perceptron
• Adaptive Propagation Graph Convolutional Network
• Improving Fairness in Graph Neural Networks via Mitigating Sensitive Attribute Leakage
• Variational Graph Normalized Auto-Encoders
• K-Core based Temporal Graph Convolutional Network for Dynamic Graphs
• MGDCF: Distance Learning via Markov Graph Diffusion for Neural Collaborative Filtering
• Should Graph Convolution Trust Neighbors? A Simple Causal Inference Method
• DropMessage: Unifying Random Dropping for Graph Neural Networks
• Multi-level Second-order Few-shot Learning
• SE-GSL: A General and Effective Graph Structure Learning Framework through Structural Entropy Optimization
• Beyond Low-pass Filtering: Graph Convolutional Networks with Automatic Filtering
• PaSca: a Graph Neural Architecture Search System under the Scalable Paradigm
• Learning Strong Graph Neural Networks with Weak Information
• Model Degradation Hinders Deep Graph Neural Networks
• Robust Mid-Pass Filtering Graph Convolutional Networks
• Differentially Private Graph Classification with GNNs
• CatGCN: Graph Convolutional Networks with Categorical Node Features
• Adaptive Dependency Learning Graph Neural Networks
• FreeKD: Free-direction Knowledge Distillation for Graph Neural Networks
• Recent Advances in Heterogeneous Relation Learning for Recommendation
• Understanding the Message Passing in Graph Neural Networks via Power Iteration Clustering
• Lifelong Learning of Graph Neural Networks for Open-World Node Classification
• GraphWorld: Fake Graphs Bring Real Insights for GNNs
• NetFense: Adversarial Defenses against Privacy Attacks on Neural Networks for Graph Data
• ROD: Reception-aware Online Distillation for Sparse Graphs
• Meta-Weight Graph Neural Network: Push the Limits Beyond Global Homophily
• Graph-adaptive Rectified Linear Unit for Graph Neural Networks
• Structure-Aware DropEdge Towards Deep Graph Convolutional Networks
• Graph Neural Networks with Diverse Spectral Filtering
• Pre-train and Learn: Preserve Global Information for Graph Neural Networks
• Transition Propagation Graph Neural Networks for Temporal Networks
• Beyond Graph Convolutional Network: An Interpretable Regularizer-centered Optimization Framework
• Predicting Information Pathways Across Online Communities
• Prior and Posterior Networks: A Survey on Evidential Deep Learning Methods For Uncertainty Estimation
• View-based Explanations for Graph Neural Networks
• Residual Network and Embedding Usage: New Tricks of Node Classification with Graph Convolutional Networks
• GraphGLOW: Universal and Generalizable Structure Learning for Graph Neural Networks
• Simple and Efficient Partial Graph Adversarial Attack: A New Perspective
• Counterfactual Learning on Graphs: A Survey
• Edge-based Local Push for Personalized PageRank
• Graph Neural Bandits
• Finding Heterophilic Neighbors via Confidence-based Subgraph Matching for Semi-supervised Node Classification
• The Impact of Global Structural Information in Graph Neural Networks Applications
• SNoRe: Scalable Unsupervised Learning of Symbolic Node Representations
• Clarify Confused Nodes via Separated Learning
• Unsupervised Neighborhood Propagation Kernel Layers for Semi-supervised Node Classification
• Graph-based Village Level Poverty Identification
• Accelerating Dynamic Network Embedding with Billions of Parameter Updates to Milliseconds
• PushNet: Efficient and Adaptive Neural Message Passing
• Boosting Multitask Learning on Graphs through Higher-Order Task Affinities
• Neural Bandit with Arm Group Graph
• DeepHGCN: Toward Deeper Hyperbolic Graph Convolutional Networks
• Scalable Deep Graph Clustering with Random-walk based Self-supervised Learning
• On the Impact of Communities on Semi-supervised Classification Using Graph Neural Networks
• Focus Where It Matters: Graph Selective State Focused Attention Networks
• Expanding Semantic Knowledge for Zero-shot Graph Embedding
• Fea2Fea: Exploring Structural Feature Correlations via Graph Neural Networks
• Classification of vertices on social networks by multiple approaches