论文标题
充血和ETA预测的分层图结构
Hierarchical Graph Structures for Congestion and ETA Prediction
论文作者
论文摘要
流量4cast是一项年度竞争，旨在根据现实世界数据预测时空流量。我们使用图形神经网络提出了一种方法，该方法直接适用于从OpenStreetMap数据中提取的道路图拓扑。我们的体系结构可以合并一个层次图表示，以改善图形的关键相交与连接它们的最短路径之间的信息流。此外，我们研究了如何压实道路图，以减轻信息流并利用多任务方法来预测拥塞类和ETA。我们的代码和模型在此处发布：https：//github.com/floriangroetschla/neurips2022-traffic4cast
Traffic4cast is an annual competition to predict spatio temporal traffic based on real world data. We propose an approach using Graph Neural Networks that directly works on the road graph topology which was extracted from OpenStreetMap data. Our architecture can incorporate a hierarchical graph representation to improve the information flow between key intersections of the graph and the shortest paths connecting them. Furthermore, we investigate how the road graph can be compacted to ease the flow of information and make use of a multi-task approach to predict congestion classes and ETA simultaneously. Our code and models are released here: https://github.com/floriangroetschla/NeurIPS2022-traffic4cast
