2020
DOI: 10.48550/arxiv.2006.11583
|View full text |Cite
Preprint
|
Sign up to set email alerts
|

A3T-GCN: Attention Temporal Graph Convolutional Network for Traffic Forecasting

Abstract: Accurate real-time traffic forecasting is a core technological problem against the implementation of the intelligent transportation system. However, it remains challenging considering the complex spatial and temporal dependencies among traffic flows. In the spatial dimension, due to the connectivity of the road network, the traffic flows between linked roads are closely related. In terms of the temporal factor, although there exists a tendency among adjacent time points in general, the importance of distant pa… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
12
0

Year Published

2021
2021
2024
2024

Publication Types

Select...
5
2

Relationship

0
7

Authors

Journals

citations
Cited by 7 publications
(12 citation statements)
references
References 22 publications
0
12
0
Order By: Relevance
“…Traffic prediction models. We evaluate the developed diffusion attack framework on three traffic prediction models: T-GCN [30], ST-GCN [31], and A3T-GCN [32], which are all based on GCN structures. For each model, we set S = 12 and T = 1.…”
Section: A Experiments Setupmentioning
confidence: 99%
“…Traffic prediction models. We evaluate the developed diffusion attack framework on three traffic prediction models: T-GCN [30], ST-GCN [31], and A3T-GCN [32], which are all based on GCN structures. For each model, we set S = 12 and T = 1.…”
Section: A Experiments Setupmentioning
confidence: 99%
“…Pareja et al [23] utilized RNN to evolve GCN parameters to capture the dynamism of the input graph sequence. Recently, the attention mechanism has been embedded into GCN to further discover the crucial temporal patterns, e.g., [41] and [12] adopt attention mechanism to assign different weights to historical information. However, it is still difficult to model long-term dependency among the high dimensional data with spatialtemporal graph convolutional network.…”
Section: A Graph Convolutional Networkmentioning
confidence: 99%
“…Moreover, Li et al [18] and Zhao et al [40] respectively proposed Diffusion Convolutional Recurrent Neural Network that combines GCN with RNN and Temporal GCN (T-GCN) that combines GCN with GRU to model spatial-temporal patterns of traffic data. Afterwards, Zhu et al [41] and Park et al [24] embedded attention mechanism into T-GCN to further capture the dynamic traffic patterns.…”
Section: B Traffic Predictionmentioning
confidence: 99%
See 2 more Smart Citations