Hyperbolic Heterogeneous Graph Transformer

TL;DR

This paper introduces HypHGT, a hyperbolic space-based transformer model for heterogeneous graphs that captures complex structures efficiently, outperforming existing methods in node classification with less computational cost.

Contribution

The paper proposes HypHGT, a novel hyperbolic transformer architecture that effectively models both local and global dependencies in heterogeneous graphs without tangent-space operations.

Findings

HypHGT outperforms state-of-the-art methods in node classification.

It achieves this with reduced training time and memory usage.

The relation-specific hyperbolic attention mechanism is efficient and preserves heterogeneous information.

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.