2018 |
DIFFUSION CONVOLUTIONAL RECURRENT NEURAL NETWORK: DATA-DRIVEN TRAFFIC FORECASTING (DCRNN) |
Volume Prediction |
Spatial: Diffusion & Graph Conv; Temporal: GRU based on Graph Conv |
ICLR |
Link |
2018 |
Spatio-Temporal Graph Convolutional Networks: A Deep Learning Framework for Traffic Forecasting(STGCN) |
Volume Prediction |
Spatial: Graph Conv; Temporal: Gated CNN |
IJCAI |
Link |
2019 |
Graph WaveNet for Deep Spatial-Temporal Graph Modeling |
Volume Prediction |
Spatial: 2 diffusion + adaptive; Temporal: gated & dilation conv |
IJCAI |
Link |
2019 |
Attention Based Spatial-Temporal Graph Convolutional Networks for Traffic Flow Forecasting(ASTGCN) |
Volume Prediction |
Spatial & Temporal: Volume to conduct attention; Add feature |
AAAI |
Link |
2020 |
Adaptive Graph Convolutional Recurrent Network for Traffic Forecasting(AGCRN) |
Volume Prediction |
Learnable pattern pool; Spatial: Embedding matrix mult; Temporal: GRU |
NIPS |
Link |
2020 |
Spatial-Temporal Synchronous Graph Convolutional Networks: A New Framework for Spatial-Temporal Network Data Forecasting(STSGCN) |
Volume Prediction |
Localized spatial-temporal graph; |
AAAI |
Link |
2021 |
MDTP: A Multi-source Deep Traffic Prediction Framework over Spatio-Temporal Trajectory Data(MDTP) |
Traffic in/out prediction |
Trajectory as edge and node feature; base GCN & LSTM |
VLDB |
Link |
2021 |
Dynamic and Multi-faceted Spatio-temporal Deep Learning for Traffic Speed Forecasting(DMSTGCN) |
Volume-auxiliaried speed pediction |
Spatial: Tucker decomposition based dynamic matrix; Temporal: dilation & gated conv; Auxiliary data |
KDD |
Link |
2021 |
Gallat: A Spatiotemporal Graph Attention Network for Passenger Demand Prediction |
Human flow prediction |
Grid map; classify 3 types of neighbor; No input feature; Spatial attention to 3 neighbors and do GraphSAGE; Temporal do attention to periodical data; predict the out degree |
ICDE short |
Link |
2021 |
Origin-Destination Matrix Prediction via Graph Convolution: a New Perspective of Passenger Demand Modeling |
OD demand prediction |
Grid map; classify 2 types of neighbor; Spatial attention to 2 neighbors and do GraphSAGE; Temporal LSTM to the output of Spatial attention; predict the OD matrix and in&out demand; Multi-task learning is tricky |
KDD |
Link |