Abstract:Graph neural networks generalize conventional neural networks to graph-structured data and have received widespread attention due to their impressive representation ability. In spite of the remarkable achievements, the performance of Euclidean models in graph-related learning is still bounded and limited by the representation ability of Euclidean geometry, especially for datasets with highly non-Euclidean latent anatomy. Recently, hyperbolic space has gained increasing popularity in processing graph data with tree-like structure and power-law distribution, owing to its exponential growth property. In this survey, we comprehensively revisit the technical details of the current hyperbolic graph neural networks, unifying them into a general framework and summarizing the variants of each component. More importantly, we present various HGNN-related applications. Last, we also identify several challenges, which potentially serve as guidelines for further flourishing the achievements of graph learning in hyperbolic spaces.

What problem does this paper attempt to address?

The paper primarily explores the research progress and applications of Hyperbolic Graph Neural Networks (HGNNs), highlighting the challenges in current research and future development directions. Specifically, the paper attempts to address the following issues: 1. **Representation capability of complex graph data**: Traditional Graph Neural Networks (GNNs) in Euclidean space have limited representation capabilities when dealing with datasets that have highly non-Euclidean latent linear structures. Due to its exponential growth characteristics, hyperbolic space performs excellently in handling tree-like structures or power-law distributed data. 2. **Review and unification of existing methods**: The paper provides a detailed technical review of existing hyperbolic graph neural network models and unifies these models through a general framework, summarizing the variants of each module. Additionally, it discusses the latest theoretical and empirical analyses of hyperbolic graph neural network models. 3. **Systematic classification of application scenarios**: The paper systematically classifies the application scenarios of hyperbolic graph neural networks and introduces several major applications and their corresponding methods for each case. These applications include recommendation systems, knowledge graphs, and drug molecules, among others. 4. **Challenges and opportunities for future research**: The paper summarizes several challenges that may be faced in future research and provides insights for further development to promote the prosperity of graph learning outcomes based on hyperbolic space. Through the research in the above aspects, the paper aims to fill the current gap in the review of hyperbolic graph neural network methods and applications and to promote further development in this field.

Hyperbolic Graph Neural Networks: A Review of Methods and Applications

Hyperbolic Graph Neural Networks: A Review of Methods and Applications

Hypergraph Learning: Methods and Practices

Hyperbolic Graph Representation Learning: A Tutorial

Hyperbolic Deep Neural Networks: A Survey

Graph Neural Network: A Comprehensive Review on Non-Euclidean Space

Graph Neural Networks: Methods, Applications, and Opportunities

Ha-gnn: a novel graph neural network based on hyperbolic attention

Neural Networks on Hypergraph

Shedding Light on Problems with Hyperbolic Graph Learning

A survey of graph neural networks in various learning paradigms: methods, applications, and challenges

Neural Embeddings of Graphs in Hyperbolic Space

A Comprehensive Survey on Graph Neural Networks

A comprehensive review of graph convolutional networks: approaches and applications

Hyperbolic Curvature Graph Neural Network

A review of graph neural networks: concepts, architectures, techniques, challenges, datasets, applications, and future directions

Graph Neural Networks: Taxonomy, Advances and Trends

Graph Neural Networks: Taxonomy, Advances, and Trends

A Survey on Hypergraph Neural Networks: An In-Depth and Step-By-Step Guide