TMetaNet: Topological Meta-Learning Framework for Dynamic Link Prediction

2025-05-31

AI Summary

Dynamic graph link prediction faces challenges arising from continuous structural evolution and strong temporal dependencies. Existing meta-learning approaches, which rely on fixed-weight updates, struggle to capture higher-order topological evolution patterns, resulting in insufficient rapid adaptation across time steps. To address this, we propose Dowker Zigzag Persistence (DZP), a novel dynamic persistent homology representation method that explicitly embeds dynamic higher-order topological distances into the meta-learning parameter update mechanism—marking the first such integration. DZP synergistically combines Dowker complexes, zigzag persistent homology, graph neural networks, and a distance-aware variant of MAML. Evaluated on multiple real-world dynamic graph datasets, DZP achieves state-of-the-art performance, significantly enhancing robustness to structural noise and cross-temporal generalization capability.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Search and Optimization: Metareasoning and Metaheuristics

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Web Mining and Content Analysis: Models for Web evolution
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Dynamic graphs evolve continuously, presenting challenges for traditional graph learning due to their changing structures and temporal dependencies. Recent advancements have shown potential in addressing these challenges by developing suitable meta-learning-based dynamic graph neural network models. However, most meta-learning approaches for dynamic graphs rely on fixed weight update parameters, neglecting the essential intrinsic complex high-order topological information of dynamically evolving graphs. We have designed Dowker Zigzag Persistence (DZP), an efficient and stable dynamic graph persistent homology representation method based on Dowker complex and zigzag persistence, to capture the high-order features of dynamic graphs. Armed with the DZP ideas, we propose TMetaNet, a new meta-learning parameter update model based on dynamic topological features. By utilizing the distances between high-order topological features, TMetaNet enables more effective adaptation across snapshots. Experiments on real-world datasets demonstrate TMetaNet's state-of-the-art performance and resilience to graph noise, illustrating its high potential for meta-learning and dynamic graph analysis. Our code is available at https://github.com/Lihaogx/TMetaNet.

Problem

• Addresses dynamic graph structural changes and temporal dependencies
• Captures high-order topological features in evolving graphs
• Improves meta-learning parameter updates using topological distances

Innovation

• Uses Dowker Zigzag Persistence for dynamic graphs
• Incorporates high-order topological features in meta-learning
• Enhances adaptation across graph snapshots with distances