STEI-PCN: an efficient pure convolutional network for traffic prediction via spatial-temporal encoding and inferring

STEI-PCN: an efficient pure convolutional network for traffic prediction via spatial-temporal encoding and inferring Kai Hu a, Zhidan Zhao b,c,, Zhifeng Hao a a Department of Mathematic, School of Mathematics and Computer Sciences, Shantou University, Shantou, 515063, Guangdong, China b Department of Computer Science, School of Mathematics and Computer Sciences, Shantou University, Shantou, 515063, Guangdong, China c Complexity Computation Laboratory, Department of Computer Science, School of Mathematics and Computer Sciences, Shantou University, Shantou, 515603, Guangdong, ChinaAbstract Traffic data exhibits complex temporal, spatial, and spatial-temporal correlations. Capturing and integrating these correlations is crucial for building accurate prediction models. Although numerous deep learning-based traffic prediction models have been developed, most of these models use either independent modules to separately extract temporal and spatial correlations or joint modules to synchronously extract them, without considering the spatial-temporal correlations. Moreover, models that consider joint spatial-temporal correlations (temporal, spatial, and spatial-temporal correlations) often encounter significant challenges in accuracy and computational efficiency which prevent such models from demonstrating the expected advantages of a joint spatial-temporal correlations architecture. To address these issues, this paper proposes an efficient pure convolutional network for traffic prediction via spatial-temporal encoding and inferring (STEI-PCN). The model introduces and designs a dynamic adjacency matrix inferring module based on absolute spatial and temporal coordinates, as well as relative spa-Corresponding author at: Department of Computer Science, School of Mathematics and Computer Sciences, Shantou University, Shantou, 515063, Guangdong, China and Complexity Computation Laboratory, Department of Computer Science, School of Mathematics and Computer Sciences, Shantou University, Shantou, 515603, Guangdong, China.