Traffic forecasting as a canonical task of multivariate time series forecasting has been a significant research topic in AI community. To address the spatio-temporal heterogeneity and non-stationarity implied in the traffic stream, in this study, we propose Spatio-Temporal Meta-Graph Learning as a novel Graph Structure Learning mechanism on spatio-temporal data. Specifically, we implement this idea into Meta-Graph Convolutional Recurrent Network (MegaCRN) by plugging the Meta-Graph Learner powered by a Meta-Node Bank into GCRN encoder-decoder. We conduct a comprehensive evaluation on two benchmark datasets (METR-LA and PEMS-BAY) and a new large-scale traffic speed dataset in which traffic incident information is contained. Our model outperformed the state-of-the-arts to a large degree on all three datasets (over 27% MAE and 34% RMSE). Besides, through a series of qualitative evaluations, we demonstrate that our model can explicitly disentangle the road links and time slots with different patterns and be robustly adaptive to any anomalous traffic situations. Codes and datasets are available at https://github.com/deepkashiwa20/MegaCRN.
翻译:作为多变时间序列预报的典型任务,交通流量预测一直是AI社区的一个重要研究课题。为了解决交通流中隐含的时空差异和非常态性,我们在本研究中提议,Spatio-Temporal Met-Graph Learning 是一个关于时空数据的新型图表结构学习机制。具体地说,我们通过将Meta-Graph学习者(MegaCRN)连接到GCRN encoder-decoder,将这一想法落实到Meta-Graph 经常网络(MegaCRN)中。此外,通过一系列定性评估,我们证明我们的模型可以明确地分解交通链和PEMS-BAY这两个基准数据集(METR-LA和PEMS-BAY)以及一个新的大型交通速度数据集,其中含有交通事故信息。我们的模型大大超越了所有三个数据集(超过27%的MAE和34%的RMEE/RMEE)的状态。此外,我们还通过一系列定性评估,我们证明,我们的模型可以明确区分道路连接路段/时间设置,并且有不同的模式和MACsmasmas 。任何坚固的模型。