APAN: Asynchronous Propagation Attention Network for Real-time Temporal Graph Embedding

Abstract

Limited by the time complexity of querying k-hop neighbors in a graph database, most graph algorithms cannot be deployed online and execute millisecond-level inference. This problem dramatically limits the potential of applying graph algorithms in certain areas, such as financial fraud detection. Therefore, we propose Asynchronous Propagation Attention Network, an asynchronous continuous time dynamic graph algorithm for real-time temporal graph embedding. Traditional graph models usually execute two serial operations: first graph computation and then model inference. We decouple model inference and graph computation step so that the heavy graph query operations will not damage the speed of model inference. Extensive experiments demonstrate that the proposed method can achieve competitive performance and 8.7 times inference speed improvement in the meantime.

In Proceedings of the 2021 International Conference on Management of Data (SIGMOD/PODS '21)

References in corpus (1)

• NetSMF: Large-Scale Network Embedding as Sparse Matrix Factorization

Cited by in corpus (8)

• The Evolution of Distributed Systems for Graph Neural Networks and their Origin in Graph Processing and Deep Learning: A Survey
• Graph Sequential Neural ODE Process for Link Prediction on Dynamic and Sparse Graphs
• Transition Propagation Graph Neural Networks for Temporal Networks
• Temporal Aggregation and Propagation Graph Neural Networks for Dynamic Representation
• MSPipe: Efficient Temporal GNN Training via Staleness-Aware Pipeline
• TIDFormer: Exploiting Temporal and Interactive Dynamics Makes A Great Dynamic Graph Transformer
• Domain-Informed Negative Sampling Strategies for Dynamic Graph Embedding in Meme Stock-Related Social Networks
• Efficient Learning-based Graph Simulation for Temporal Graphs