• Corpus ID: 245650514

A Lightweight and Accurate Spatial-Temporal Transformer for Traffic Forecasting

  title={A Lightweight and Accurate Spatial-Temporal Transformer for Traffic Forecasting},
  author={Guanyao Li and Shuhan Zhong and Letian Xiang and Shueng-Han Gary Chan and Ruiyuan Li and Chih-Chieh Hung and Wen-Chih Peng},
—We study the forecasting problem for traffic with dynamic, possibly periodical, and joint spatial-temporal dependency between regions. Given the aggregated inflow and outflow traffic of regions in a city from time slots 0 to t − 1 , we predict the traffic at time t at any region. Prior arts in the area often consider the spatial and temporal dependencies in a decoupled manner, or are rather computationally intensive in training with a large number of hyper-parameters to tune. We propose ST-TIS, a… 



Revisiting Spatial-Temporal Similarity: A Deep Learning Framework for Traffic Prediction

A novel Spatial-Temporal Dynamic Network (STDN) is proposed, in which a flow gating mechanism is introduced to learn the dynamic similarity between locations, and a periodically shifted attention mechanism is designed to handle long-term periodic temporal shifting.

Spatial-Temporal Transformer Networks for Traffic Flow Forecasting

A novel paradigm of Spatial-Temporal Transformer Networks (STTNs) that leverages dynamical directed spatial dependencies and long-range temporal dependencies to improve the accuracy of long-term traffic forecasting and achieves competitive results compared with the state-of-the-arts.

DNN-based prediction model for spatio-temporal data

A Deep-learning-based prediction model for Spatio-Temporal data (DeepST), which is comprised of two components: spatio-temporal and global, and built on a real-time crowd flow forecasting system called UrbanFlow1.

Urban Traffic Prediction from Spatio-Temporal Data Using Deep Meta Learning

This work proposed a deep-meta-learning based model, entitled ST-MetaNet, to collectively predict traffic in all location at once, consisting of a recurrent neural network to encode the traffic, a meta graph attention network to capture diverse spatial correlations, and a meta recurrent Neural network to consider diverse temporal correlations.

Deep Spatio-Temporal Residual Networks for Citywide Crowd Flows Prediction

This work designs an end-to-end structure of ST-ResNet, a deep-learning-based approach to collectively forecast the inflow and outflow of crowds in each and every region of a city, based on unique properties of spatio-temporal data.

Spatio-Temporal Graph Convolutional Networks: A Deep Learning Framework for Traffic Forecasting

This paper proposes a novel deep learning framework, Spatio-Temporal Graph Convolutional Networks (STGCN), to tackle the time series prediction problem in traffic domain, and builds the model with complete convolutional structures, which enable much faster training speed with fewer parameters.

Forecaster: A Graph Transformer for Forecasting Spatial and Time-Dependent Data

The proposed Forecaster, a graph Transformer architecture, sparsify the Transformer to account for the strength of spatial dependency, long-range temporal dependency, data non-stationarity, and data heterogeneity, and is evaluated in the problem of forecasting taxi ride-hailing demand.

An Effective Joint Prediction Model for Travel Demands and Traffic Flows

A comprehensive neural network based traffic prediction model, where various effective embeddings or encodings are designed to capture three desired properties, namely region-level correlations, temporal periodicity and inter-traffic correlations is proposed.

Spatial-Temporal Fusion Graph Neural Networks for Traffic Flow Forecasting

A data-driven method of generating “temporal graph” is proposed to compensate several genuine correlations that spatial graph may not reflect, and a novel Spatial-Temporal Fusion Graph Neural Networks (STFGNN) is proposed for traffic flow forecasting.

Spatial-Temporal Synchronous Graph Convolutional Networks: A New Framework for Spatial-Temporal Network Data Forecasting

A novel model, named Spatial-Temporal Synchronous Graph Convolutional Networks (STSGCN), is proposed, which is able to effectively capture the complex localized spatial-temporal correlations through an elaborately designed spatial- Temporal synchronous modeling mechanism.