Neural Embeddings of Graphs in Hyperbolic Space

Abstract

Neural embeddings have been used with great success in Natural Language Processing (NLP). They provide compact representations that encapsulate word similarity and attain state-of-the-art performance in a range of linguistic tasks. The success of neural embeddings has prompted significant amounts of research into applications in domains other than language. One such domain is graph-structured data, where embeddings of vertices can be learned that encapsulate vertex similarity and improve performance on tasks including edge prediction and vertex labelling. For both NLP and graph based tasks, embeddings have been learned in high-dimensional Euclidean spaces. However, recent work has shown that the appropriate isometric space for embedding complex networks is not the flat Euclidean space, but negatively curved, hyperbolic space. We present a new concept that exploits these recent insights and propose learning neural embeddings of graphs in hyperbolic space. We provide experimental evidence that embedding graphs in their natural geometry significantly improves performance on downstream tasks for several real-world public datasets.

7 pages, 5 figures

References in corpus (3)

• Finding community structure in networks using the eigenvectors of matrices
• Sustaining the Internet with Hyperbolic Mapping
• Customer Lifetime Value Prediction Using Embeddings

Cited by in corpus (27)

• The Geometry of Culture: Analyzing Meaning through Word Embeddings
• Hyperbolic Graph Convolutional Neural Networks
• Machine Learning on Graphs: A Model and Comprehensive Taxonomy
• Representation Learning for Dynamic Graphs: A Survey
• Hyperbolic Attention Networks
• Skip-gram word embeddings in hyperbolic space
• Gradient descent in hyperbolic space
• Poincaré Wasserstein Autoencoder
• Using Embeddings to Correct for Unobserved Confounding in Networks
• Beltrami Flow and Neural Diffusion on Graphs
• Hyperbolic Deep Neural Networks: A Survey
• Adversarial Classifier for Imbalanced Problems
• Latent Variable Modelling with Hyperbolic Normalizing Flows
• Benchmarking neural embeddings for link prediction in knowledge graphs under semantic and structural changes
• Incorporating Domain Knowledge into Health Recommender Systems using Hyperbolic Embeddings
• Variational Integrator Networks for Physically Structured Embeddings
• Global and local evaluation of link prediction tasks with neural embeddings
• Semi-Supervised Learning on Graphs Based on Local Label Distributions
• APo-VAE: Text Generation in Hyperbolic Space
• Social Connection Induces Cultural Contraction: Evidence from Hyperbolic Embeddings of Social and Semantic Networks
• Unsupervised Hierarchy Matching with Optimal Transport over Hyperbolic Spaces
• HEAT: Hyperbolic Embedding of Attributed Networks
• Network Representation Learning: From Traditional Feature Learning to Deep Learning
• Temporal Graph Offset Reconstruction: Towards Temporally Robust Graph Representation Learning
• Nearest Neighbor Search for Hyperbolic Embeddings
• What Remains of Visual Semantic Embeddings
• Understanding in Artificial Intelligence