Continuous Temporal Graph Neural Networks

Zulun Zhu, Siqiang Luo

Keywords: Temporal graph, Graph Neural Networks, attention mechanism

Abstract:

Temporal Graph Neural Networks (TGNNs) are pivotal in processing dynamic graphs. However, existing TGNNs primarily target one-time predictions for a given temporal span, whereas many practical applications require {\it continuous predictions}, that predictions are issued frequently over time.
Directly adapting existing TGNNs to continuous-prediction scenarios introduces either significant computational overhead or prediction quality issues especially for large graphs.
This paper revisits the challenge of {\it continuous predictions} in TGNNs, and introduces {\sc Coden}, a TGNN model designed for efficient and effective learning on dynamic graphs.
{\sc Coden} innovatively overcomes the key complexity bottleneck in existing TGNNs while preserving comparable predictive accuracy. \textcolor{blue}{Moreover, we further provide theoretical analyses that substantiate the effectiveness and efficiency of {\sc Coden}, and clarify both its relationship to and improvements over attention-based models. }
Our evaluations across five dynamic datasets show that {\sc Coden} surpasses existing performance benchmarks in both efficiency and effectiveness, establishing it as a superior solution for continuous prediction in evolving graph environments.

Primary Area: learning on graphs and other geometries & topologies

Submission Number: 9418