Higher-order Structure Boosts Link Prediction on Temporal Graphs

2025-05-21

AI Summary

Existing temporal graph neural networks (TGNNs) are limited to modeling pairwise interactions, neglecting higher-order structures (e.g., group-level dependencies), and suffer from poor memory efficiency—hindering both expressive power and scalability. To address these limitations, we propose HTGN—the first temporal graph neural network integrating hypergraph representation learning. HTGN introduces higher-order structural modeling into dynamic link prediction by constructing hyperedge representations via multi-feature edge aggregation and designing a temporal-aware hypergraph message-passing mechanism. We theoretically establish that HTGN possesses strictly greater expressive power than conventional TGNNs. Empirically, HTGN achieves significant improvements in link prediction accuracy across multiple real-world temporal graph datasets. Moreover, it reduces peak training memory consumption by up to 50%, effectively balancing high expressivity with computational efficiency.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Planning, Routing, and Scheduling: Temporal Planning

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Social Networks and Social Media: Influence propagation, information diffusion, and the prediction on networks

Abstract

Temporal Graph Neural Networks (TGNNs) have gained growing attention for modeling and predicting structures in temporal graphs. However, existing TGNNs primarily focus on pairwise interactions while overlooking higher-order structures that are integral to link formation and evolution in real-world temporal graphs. Meanwhile, these models often suffer from efficiency bottlenecks, further limiting their expressive power. To tackle these challenges, we propose a Higher-order structure Temporal Graph Neural Network, which incorporates hypergraph representations into temporal graph learning. In particular, we develop an algorithm to identify the underlying higher-order structures, enhancing the model's ability to capture the group interactions. Furthermore, by aggregating multiple edge features into hyperedge representations, HTGN effectively reduces memory cost during training. We theoretically demonstrate the enhanced expressiveness of our approach and validate its effectiveness and efficiency through extensive experiments on various real-world temporal graphs. Experimental results show that HTGN achieves superior performance on dynamic link prediction while reducing memory costs by up to 50% compared to existing methods.

Problem

• Existing TGNNs overlook higher-order structures in temporal graphs
• Current models face efficiency bottlenecks limiting expressive power
• Need to enhance link prediction while reducing memory costs

Innovation

• Incorporates hypergraph representations into temporal learning
• Identifies higher-order structures to capture group interactions
• Reduces memory cost by aggregating edge features