Measuring and Improving the Use of Graph Information in Graph Neural Networks

Abstract

Graph neural networks (GNNs) have been widely used for representation learning on graph data. However, there is limited understanding on how much performance GNNs actually gain from graph data. This paper introduces a context-surrounding GNN framework and proposes two smoothness metrics to measure the quantity and quality of information obtained from graph data. A new GNN model, called CS-GNN, is then designed to improve the use of graph information based on the smoothness values of a graph. CS-GNN is shown to achieve better performance than existing methods in different types of real graphs.

arXiv commentThis paper has been published in ICLR 2020. Code and Dataset can be found here: https://github.com/yifan-h/CS-GNN

References in corpus (10)

• Distributed Representations of Words and Phrases and their Compositionality
• Community detection in graphs
• Geometric deep learning: going beyond Euclidean data
• Graph Embedding Techniques, Applications, and Performance: A Survey
• A Review of Relational Machine Learning for Knowledge Graphs
• Graph Neural Networks: A Review of Methods and Applications
• struc2vec: Learning Node Representations from Structural Identity
• Variational Graph Auto-Encoders
• Learning Structural Node Embeddings Via Diffusion Wavelets
• Utilizing Edge Features in Graph Neural Networks via Variational Information Maximization

Cited by in corpus (20)

• Enhancing Graph Neural Network-based Fraud Detectors against Camouflaged Fraudsters
• Beyond Homophily in Graph Neural Networks: Current Limitations and Effective Designs
• How to Find Your Friendly Neighborhood: Graph Attention Design with Self-Supervision
• Graph Information Bottleneck
• Is Homophily a Necessity for Graph Neural Networks?
• Is Heterophily A Real Nightmare For Graph Neural Networks To Do Node Classification?
• Graph neural network-based fault diagnosis: a review
• Subgroup Generalization and Fairness of Graph Neural Networks
• Geometer: Graph Few-Shot Class-Incremental Learning via Prototype Representation
• Spatial Heterophily Aware Graph Neural Networks
• Graph Representation Learning Beyond Node and Homophily
• Bipartite Graph Network with Adaptive Message Passing for Unbiased Scene Graph Generation
• CommPOOL: An Interpretable Graph Pooling Framework for Hierarchical Graph Representation Learning
• PSGR: Pixel-wise Sparse Graph Reasoning for COVID-19 Pneumonia Segmentation in CT Images
• Neuron with Steady Response Leads to Better Generalization
• LEReg: Empower Graph Neural Networks with Local Energy Regularization
• Graph Self-Supervised Learning with Learnable Structural and Positional Encodings
• Eigen-GNN: A Graph Structure Preserving Plug-in for GNNs
• Alleviating Over-Smoothing via Aggregation over Compact Manifolds
• CoRGi: Content-Rich Graph Neural Networks with Attention