Traffic Forecasting
Traffic forecasting aims to predict future traffic conditions using historical data and potentially external factors, enabling efficient resource allocation and improved transportation management. Current research heavily utilizes deep learning, focusing on graph neural networks (GNNs) and transformers, often combined to leverage both spatial and temporal dependencies within road networks, with a growing emphasis on handling data heterogeneity, out-of-distribution scenarios, and limited sensor coverage. These advancements hold significant potential for improving urban planning, optimizing traffic flow, and enhancing the efficiency of intelligent transportation systems.
Papers
STGformer: Efficient Spatiotemporal Graph Transformer for Traffic Forecasting
Hongjun Wang, Jiyuan Chen, Tong Pan, Zheng Dong, Lingyu Zhang, Renhe Jiang, Xuan Song
Robust Traffic Forecasting against Spatial Shift over Years
Hongjun Wang, Jiyuan Chen, Tong Pan, Zheng Dong, Lingyu Zhang, Renhe Jiang, Xuan Song