融合CNN与LSTM的时空图网络新架构，提升时序建模能力。
Spatio-Temporal Graph Convolutional Networks: Optimised Temporal Architecture
- 设计混合CNN-LSTM时序模块，兼顾局部特征与长程依赖。
- 在多个数据集上验证性能优于单一结构，提升显著。
- 适合处理具有复杂时空动态的轨迹预测与行为分析任务。
时空图卷积网络最初采用卷积神经网络（CNN）作为时序模块进行特征提取。此后，基于长短期记忆网络（LSTM）的时序模块被提出并展现出良好效果。本文提出一种结合CNN与LSTM时序模块的新架构，并对新模型与现有模型进行实证比较。通过理论分析不同时间模块的特性，并在多个数据集上开展多组测试，全面评估假设。结果表明，混合架构在多种场景下表现更优，验证了联合利用两者优势的有效性。
原文摘要 · Abstract (English)
Spatio-Temporal graph convolutional networks were originally introduced with CNNs as temporal blocks for feature extraction. Since then LSTM temporal blocks have been proposed and shown to have promising results. We propose a novel architecture combining both CNN and LSTM temporal blocks and then provide an empirical comparison between our new and the pre-existing models. We provide theoretical arguments for the different temporal blocks and use a multitude of tests across different datasets to assess our hypotheses.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。