TMetaNet: Topological Meta-Learning Framework for Dynamic Link Prediction

Hao Li, Hao Wan, Yuzhou Chen, Dongsheng Ye, Yulia Gel, Hao Jiang
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:35977-35999, 2025.

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.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-li25ci, title = {{TM}eta{N}et: Topological Meta-Learning Framework for Dynamic Link Prediction}, author = {Li, Hao and Wan, Hao and Chen, Yuzhou and Ye, Dongsheng and Gel, Yulia and Jiang, Hao}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {35977--35999}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/li25ci/li25ci.pdf}, url = {https://proceedings.mlr.press/v267/li25ci.html}, 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.} }
Endnote
%0 Conference Paper %T TMetaNet: Topological Meta-Learning Framework for Dynamic Link Prediction %A Hao Li %A Hao Wan %A Yuzhou Chen %A Dongsheng Ye %A Yulia Gel %A Hao Jiang %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-li25ci %I PMLR %P 35977--35999 %U https://proceedings.mlr.press/v267/li25ci.html %V 267 %X 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.
APA
Li, H., Wan, H., Chen, Y., Ye, D., Gel, Y. & Jiang, H.. (2025). TMetaNet: Topological Meta-Learning Framework for Dynamic Link Prediction. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:35977-35999 Available from https://proceedings.mlr.press/v267/li25ci.html.

Related Material