Feb 6, 2024 · This paper studies the efficient training of MDGNNs at scale, focusing on the temporal discontinuity in training MDGNNs with large temporal batch sizes.

Memory-based Dynamic Graph Neural Networks (MDGNNs) are a family of dy- namic graph neural networks that leverage a memory module to extract, distill,.

PRES: Toward Scalable Memory-Based Dynamic Graph Neural Networks. Open Webpage · Junwei Su, Difan Zou, Chuan Wu. Published: 31 Dec 2023, Last Modified: 30 ...

Feb 26, 2024 · Among DGNNs, Memory-based Dynamic Graph Neural Networks (MDGNNs) have demonstrated superior performance compared to memory-less counterparts ...

International Conference on Learning Representations (ICLR) (Spotlight), 2024 [Paper][Arxiv]. PRES: Toward Scalable Memory-Based Dynamic Graph Neural Networks

Awesome papers (codes) about machine learning (deep learning) on dynamic (temporal) graphs (networks / knowledge graphs) and their applications.

PRES: Toward Scalable Memory-Based Dynamic Graph Neural Networks, Junwei Su, Difan Zou, Chuan Wu, here. One For All: Towards Training One Graph Model For All ...

Memory-based Dynamic Graph Neural Networks (MDGNNs) are a family of dynamic graph neural networks that leverage a memory module to extract, distill, and ...

PRES: Toward Scalable Memory-Based Dynamic Graph Neural Networks · no code implementations • 6 Feb 2024 • Junwei Su, Difan Zou, Chuan Wu. Memory-based Dynamic ...

International Conference on Machine Learning, 32728-32748, 2023. 14, 2023. PRES: Toward Scalable Memory-Based Dynamic Graph Neural Networks. J Su, D Zou, C Wu.