Hyperbolic Heterogeneous Graph Transformer
Jongmin Park, Seunghoon Han, Hyewon Lee, Won-Yong Shin, Sungsu Lim
TL;DR
HypHGT addresses the challenge of learning representations for heterogeneous graphs with hierarchical structures by learning entirely in hyperbolic space and employing a relation-aware hyperbolic transformer. It combines a linear-time hyperbolic attention mechanism operating in relation-specific spaces with heterogeneous GNNs to capture global and local information, then fuses these signals for final node embeddings. Empirical results across real and synthetic datasets show state-of-the-art node classification performance with significantly improved efficiency, and analyses reveal adaptive curvature learning that aligns with relation-specific hierarchies. The work provides a scalable, principled approach to modeling complex heterogeneous graphs and suggests promising avenues for dynamic and cross-domain extensions.
Abstract
In heterogeneous graphs, we can observe complex structures such as tree-like or hierarchical structures. Recently, the hyperbolic space has been widely adopted in many studies to effectively learn these complex structures. Although these methods have demonstrated the advantages of the hyperbolic space in learning heterogeneous graphs, most existing methods still have several challenges. They rely heavily on tangent-space operations, which often lead to mapping distortions during frequent transitions. Moreover, their message-passing architectures mainly focus on local neighborhood information, making it difficult to capture global hierarchical structures and long-range dependencies between different types of nodes. To address these limitations, we propose Hyperbolic Heterogeneous Graph Transformer (HypHGT), which effectively and efficiently learns heterogeneous graph representations entirely within the hyperbolic space. Unlike previous message-passing based hyperbolic heterogeneous GNNs, HypHGT naturally captures both local and global dependencies through transformer-based architecture. Furthermore, the proposed relation-specific hyperbolic attention mechanism in HypHGT, which operates with linear time complexity, enables efficient computation while preserving the heterogeneous information across different relation types. This design allows HypHGT to effectively capture the complex structural properties and semantic information inherent in heterogeneous graphs. We conduct comprehensive experiments to evaluate the effectiveness and efficiency of HypHGT, and the results demonstrate that it consistently outperforms state-of-the-art methods in node classification task, with significantly reduced training time and memory usage.