ST-MGAT: Spatial-Temporal Multi-Head Graph Attention Networks for Traffic Forecasting

Tian, Kelang; Guo, Jingjie; Ye, Kejiang; Xu, Cheng‐Zhong

doi:10.1109/ictai50040.2020.00114

Cited by 7 publications

(5 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A machine learning method for regression problems. FC‐LSTM [50]: Long short‐term memory network, a sequence‐to‐sequence model based on the recurrent neural network with fully‐connected LSTM hidden units. DCRNN [49]: Diffusion convolutional recurrent neural network, which uses diffusion graph convolutional networks and the encoder‐decoder structure to capture the spatial and temporal dependencies, respectively. STGCN [6]: Spatio‐temporal graph convolution networks, which integrate the gated temporal convolution unit into graph convolution blocks. GraphWaveNet [17]: Graph WaveNet combines an adaptive adjacency matrix into graph convolution and utilizes stacked 1D convolution units to capture the spatial‐temporal dependency. STSGCN [14]: Spatial‐temporal synchronous graph convolutional networks, which effectively capture the localized and long‐range spatial‐temporal dependencies through a spatial‐temporal synchronous modeling mechanism. GMAN [9]: Graph multi‐attention network, an encoder‐decoder architecture with multiple spatial‐temporal attention blocks. The transform attention layers enable modeling the impact of the spatio‐temporal factors on complex traffic conditions. ST‐MGAT [51]: Spatial‐temporal multi‐head graph attention network, which consists of temporal convolution blocks and graph attention networks for capturing the dynamic temporal correlations and spatial relations between nodes, respectively.…”

Section: Methodsmentioning

confidence: 99%

“…The transform attention layers enable modeling the impact of the spatio-temporal factors on complex traffic conditions. • ST-MGAT [51]: Spatial-temporal multi-head graph attention network, which consists of temporal convolution blocks and graph attention networks for capturing the dynamic temporal correlations and spatial relations between nodes, respectively.…”

Section: Baseline Descriptionmentioning

confidence: 99%

See 1 more Smart Citation

Spatial‐temporal correlation graph convolutional networks for traffic forecasting

Huang

Chen

Zhai

et al. 2023

IET Intelligent Trans Sys

View full text Add to dashboard Cite

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Baseline Descriptionmentioning

confidence: 99%

Spatial‐temporal correlation graph convolutional networks for traffic forecasting

Huang

Chen

Zhai

et al. 2023

IET Intelligent Trans Sys

View full text Add to dashboard Cite

show abstract

“…Using a matrix factorization technique, Bai et al [33] suggested a convolutional GNN module that can apply node specific parameters. Attentional GNNs also have been widely used in traffic forecasting research [21], [26], [27], [36], [41], [45], [52], [66], [73]. The gated attention networks (GaAN) [26] outperformed diffusion convolution in short-term traffic forecasting when combined with GRU.…”

Section: B Spatial Feature Extraction With Graph Neural Networkmentioning

confidence: 99%

“…The gated attention networks (GaAN) [26] outperformed diffusion convolution in short-term traffic forecasting when combined with GRU. GAT [79] has also been adopted in many studies [21], [27], [36], [41], [52], [73]. Park et al [66] constructed a new attentional GNN layer that adopts the scaled dot-product attention [61] with sentinel vectors to control the information from neighbor nodes.…”

Section: B Spatial Feature Extraction With Graph Neural Networkmentioning

confidence: 99%

Performance Evaluation of Building Blocks of Spatial-Temporal Deep Learning Models for Traffic Forecasting

Shin,

Yoon

2023

IEEE Access

View full text Add to dashboard Cite

The traffic forecasting problem is a challenging task that requires spatial-temporal modeling and gathers research interests from various domains. In recent years, spatial-temporal deep learning models have improved the accuracy and scale of traffic forecasting. While hundreds of models have been suggested, they share similar modules, or building blocks, which can be categorized into three temporal feature extraction methods of recurrent neural networks, convolution, and self-attention and two spatial feature extraction methods of convolutional graph neural networks (GNN) and attentional GNN. More importantly, the models have been mostly evaluated for their entire architectures with limited efforts to characterize and understand the performance of each category of building blocks. In this study, we conduct an extensive, multifaceted experiment to understand the influence of building block selection on traffic forecasting accuracy, considering environmental characteristics and dataset distributions. Specifically, we implement six traffic forecasting models using three temporal and two spatial building blocks. When we evaluate the models on four datasets with diverse characteristics, the results show each building block demonstrates distinguishable characteristics depending on study sites, prediction horizons, and traffic categories. The convolution models demonstrate higher overall forecasting performance than other models, whereas self-attention models show competitiveness in less frequent traffic categories, transition states, and the presence of outliers. Based on the results, we also suggest an adaptive model evaluation framework for category-wise predictions of test sets based on the performance of the models on validation sets. The results of this evaluation framework demonstrate improved forecasting accuracy at most by 3.7% without further sophistication in existing model architectures. The results enhance the utility of existing models and suggest guidelines for researchers building traffic forecasting model architectures and for practitioners implementing these state-of-the-art techniques in real-world applications.

show abstract

“…STGCN [18], AGCRN [35], and the proposed model GAGW in this paper belong to spatiotemporal graph models. On the other hand, based on different time layer processing strategies, models can be further categorized into RNN-based prediction models led by RNN and DCRNN [21], TCN models such as GWNET [36] and STMGAT [37], and self-attention-mechanism-based models such as STTN [38].…”

Section: Baseline and Metricmentioning

confidence: 99%

Research on Multi-Port Ship Traffic Prediction Method Based on Spatiotemporal Graph Neural Networks

Mei

et al. 2023

JMSE

View full text Add to dashboard Cite

The intelligent maritime transportation system has emerged as a pivotal component in port management, owing to the rapid advancements in artificial intelligence and big data technology. Its essence lies in the application of digital modeling techniques, which leverage extensive ship data to facilitate efficient operations. In this regard, effective modeling and accurate prediction of the fluctuation patterns of ship traffic in multiple port regions will provide data support for trade analysis, port construction planning, and traffic safety management. In order to better express the potential interdependencies between ports, inspired by graph neural networks, this paper proposes a data-driven approach to construct a multi-port network and designs a spatiotemporal graph neural network model. The model incorporates graph attention networks and a dilated causal convolutional architecture to capture the temporal and spatial dimensions of traffic variation patterns. It also employs a gated-mechanism-based spatiotemporal bi-dimensional feature fusion strategy to handle the potential unequal relationships between the two dimensions of features. Compared to existing methods for port traffic prediction, this model fully considers the network characteristics of the overall port and fills the research gap in multi-port scenarios. In the experiments, real port ship traffic datasets were constructed using data from the Automatic Identification System (AIS) and port geographical information data for model validation. The results demonstrate that the model exhibits outstanding robustness and performs well in predicting traffic in multiple sub-regional port clusters.

show abstract

ST-MGAT: Spatial-Temporal Multi-Head Graph Attention Networks for Traffic Forecasting

Cited by 7 publications

References 6 publications

Spatial‐temporal correlation graph convolutional networks for traffic forecasting

Spatial‐temporal correlation graph convolutional networks for traffic forecasting

Performance Evaluation of Building Blocks of Spatial-Temporal Deep Learning Models for Traffic Forecasting

Research on Multi-Port Ship Traffic Prediction Method Based on Spatiotemporal Graph Neural Networks

Contact Info

Product

Resources

About