Publication Type
Journal Article
Version
submittedVersion
Publication Date
9-2022
Abstract
Graph Neural Networks (GNNs) have demonstrated state-of-the-art performance in a wide variety of analytical tasks. Current GNN approaches focus on learning representations in a Euclidean space, which are effective in capturing non-tree-like structural relations, but they fail to model complex relations in many real-world graphs, such as tree-like hierarchical graph structure. This paper instead proposes to learn representations in both Euclidean and hyperbolic spaces to model these two types of graph geometries. To this end, we introduce a novel approach - Joint hyperbolic and Euclidean geometry contrastive graph neural networks (JointGMC). JointGMC is enforced to learn multiple layer-wise optimal combinations of Euclidean and hyperbolic geometries to effectively encode diverse complex graph structures. Further, the performance of most GNNs relies heavily on the availability of large-scale manually labeled data. To mitigate this issue, JointGMC exploits proximitybased self-supervised information in different geometric spaces (i.e., Euclidean, hyperbolic, and Euclidean-hyperbolic interaction spaces) to regularize the (semi-) supervised graph learning. Extensive experimental results on eight real-world graph datasets show that JointGMC outperforms eight state-of-the-art GNN models in diverse graph mining tasks, including node classification, link prediction, and node clustering tasks, demonstrating JointGMC's superior graph representation ability. Code is available at https://github.com/chachatang/jointgmc. (c) 2022 Elsevier Inc. All rights reserved.
Keywords
Graph neural networks, Hyperbolic embedding, Contrastive learning, Graph representation learning
Discipline
Databases and Information Systems | Graphics and Human Computer Interfaces | OS and Networks
Research Areas
Data Science and Engineering
Publication
Information Sciences
Volume
609
First Page
799
Last Page
815
ISSN
0020-0255
Identifier
10.1016/j.ins.2022.07.060
Publisher
Elsevier
Citation
XU, Xiaoyu; PANG, Guansong; WU, Di; and SHANG, Mingsheng.
Joint hyperbolic and Euclidean geometry contrastive graph neural networks. (2022). Information Sciences. 609, 799-815.
Available at: https://ink.library.smu.edu.sg/sis_research/7564
Additional URL
http://doi.org/10.1016/j.ins.2022.07.060
Included in
Databases and Information Systems Commons, Graphics and Human Computer Interfaces Commons, OS and Networks Commons