Hyperbolic Graph Convolutional Neural Networks

Abstract

Graph convolutional neural networks (GCNs) embed nodes in a graph into Euclidean space, which has been shown to incur a large distortion when embedding real-world graphs with scale-free or hierarchical structure. Hyperbolic geometry offers an exciting alternative, as it enables embeddings with much smaller distortion. However, extending GCNs to hyperbolic geometry presents several unique challenges because it is not clear how to define neural network operations, such as feature transformation and aggregation, in hyperbolic space. Furthermore, since input features are often Euclidean, it is unclear how to transform the features into hyperbolic embeddings with the right amount of curvature. Here we propose Hyperbolic Graph Convolutional Neural Network (HGCN), the first inductive hyperbolic GCN that leverages both the expressiveness of GCNs and hyperbolic geometry to learn inductive node representations for hierarchical and scale-free graphs. We derive GCN operations in the hyperboloid model of hyperbolic space and map Euclidean input features to embeddings in hyperbolic spaces with different trainable curvature at each layer. Experiments demonstrate that HGCN learns embeddings that preserve hierarchical structure, and leads to improved performance when compared to Euclidean analogs, even with very low dimensional embeddings: compared to state-of-the-art GCNs, HGCN achieves an error reduction of up to 63.1% in ROC AUC for link prediction and of up to 47.5% in F1 score for node classification, also improving state-of-the art on the Pubmed dataset.

Published at Conference NeurIPS 2019. First 2 authors have equal contribution

References in corpus (9)

• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• Inductive Representation Learning on Large Graphs
• Hierarchical structure and the prediction of missing links in networks
• Simplifying Graph Convolutional Networks
• Variational Graph Auto-Encoders
• The Large Scale Curvature of Networks
• Riemannian SVRG: Fast Stochastic Optimization on Riemannian Manifolds
• Hyperbolic Graph Neural Networks
• Neural Embeddings of Graphs in Hyperbolic Space

Cited by in corpus (31)

• Network Geometry
• Differentiable Graph Module (DGM) for Graph Convolutional Networks
• Graph Convolutional Networks for Graphs Containing Missing Features
• On Proximity and Structural Role-based Embeddings in Networks: Misconceptions, Techniques, and Applications
• Labeling Trick: A Theory of Using Graph Neural Networks for Multi-Node Representation Learning
• Dual Space Graph Contrastive Learning
• HICF: Hyperbolic Informative Collaborative Filtering
• Graph Neural Networks on SPD Manifolds for Motor Imagery Classification: A Perspective from the Time-Frequency Analysis
• Self-Supervised Hyperboloid Representations from Logical Queries over Knowledge Graphs
• Where are we in embedding spaces? A Comprehensive Analysis on Network Embedding Approaches for Recommender Systems
• Ultrahyperbolic Knowledge Graph Embeddings
• Graph Geometry Interaction Learning
• Hyperbolic Geometric Graph Representation Learning for Hierarchy-imbalance Node Classification
• Bilinear Graph Neural Network with Neighbor Interactions
• Parameterized Hypercomplex Graph Neural Networks for Graph Classification
• Heterogeneous Social Event Detection via Hyperbolic Graph Representations
• BSAL: A Framework of Bi-component Structure and Attribute Learning for Link Prediction
• Size-Invariant Graph Representations for Graph Classification Extrapolations
• Generating Topological Structure of Floorplans from Room Attributes
• Hyperbolic Deep Learning for Foundation Models: A Survey
• Reconstruction for Powerful Graph Representations
• DeepHGCN: Toward Deeper Hyperbolic Graph Convolutional Networks
• Local topological moves determine global diffusion properties of hyperbolic higher-order networks
• Learning Hyperbolic Representations of Topological Features
• Single-Layer Graph Convolutional Networks For Recommendation
• A Geometric Perspective for High-Dimensional Multiplex Graphs
• ${\text{H}}^2\text{TNE}$: Temporal Heterogeneous Information Network Embedding in Hyperbolic Spaces
• A Theory of Hyperbolic Prototype Learning
• Weisfeiler and Leman go Hyperbolic: Learning Distance Preserving Node Representations
• A Survey on Graph Neural Networks for Fraud Detection in Ride Hailing Platforms
• A Literature Review of Recent Graph Embedding Techniques for Biomedical Data