APAN: Asynchronous Propagation Attention Network for Real-time Temporal Graph Embedding
Authors:
- Xuhong WangXWXuhong Wang
- Ding LyuDLDing Lyu
- Mengjian LiMLMengjian Li
- Yang XiaYXYang Xia
- Qi YangQYQi Yang
- Xinwen WangXWXinwen Wang
- Xinguang WangXWXinguang Wang
- Ping CuiPCPing Cui
- Yupu YangYYYupu Yang
- Bowen SunBSBowen Sun
- Zhenyu GuoZGZhenyu Guo
Preprint
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.Keywords
All Related Versions
This publication has 9 references indexed in Scilit:
- Learning Temporal Interaction Graph Embedding via Coupled Memory Networks2020
- E-LSTM-D: A Deep Learning Framework for Dynamic Network Link PredictionIEEE Transactions on Systems, Man, and Cybernetics: Systems, 2019
- Predicting Dynamic Embedding Trajectory in Temporal Interaction Networks2019
- Embedding Temporal Network via Neighborhood Formation2018
- Graph Convolutional Neural Networks for Web-Scale Recommender Systems2018
- Subgraph Pattern Neural Networks for High-Order Graph Evolution PredictionProceedings of the AAAI Conference on Artificial Intelligence, 2018
- Dynamic Network Embedding by Modeling Triadic Closure ProcessProceedings of the AAAI Conference on Artificial Intelligence, 2018
- Community Preserving Network EmbeddingProceedings of the AAAI Conference on Artificial Intelligence, 2017
- Structural Deep Network Embedding2016