Robust Knowledge Adaptation for Dynamic Graph Neural Networks

arXiv:2207.10839v2 Announce Type: replace
Abstract: Graph structured data often possess dynamic characters in nature. Recent years have witnessed the increasing attentions paid to dynamic graph neural networks for modelling graph data. However, almost all existing approaches operate under the assumption that, upon the establishment of a new link, the embeddings of the neighboring nodes should undergo updates to learn temporal dynamics. Nevertheless, these approaches face the following limitation: If the node introduced by a new connection contains noisy information, propagating …

abstract arxiv characters cs.lg data dynamic embeddings graph graph data graph neural networks however knowledge modelling nature networks neural networks robust structured data type