Graph WaveNet for Deep Spatial-Temporal Graph Modeling
By Zonghan Wu, Shirui Pan, Guodong Long, Jing Jiang, Chengqi Zhang. International Joint Conference on Artificial Intelligence, 2019
3,288 citations recorded by the source corpus.
Spatial-temporal graph modeling is an important task to analyze the spatial relations and temporal trends of components in a system. Existing approaches mostly capture the spatial dependency on a fixed graph structure, assuming that the underlying relation between entities is pre-determined. However, the explicit graph structure (relation) does not necessarily reflect the true dependency and genuine relation may be missing due to the incomplete connections in the data. Furthermore, existing methods are ineffective to capture the temporal trends as the RNNs or CNNs employed in these methods cannot capture long-range temporal sequences. To overcome these limitations, we propose in this paper a novel graph neural network architecture, {Graph WaveNet}, for spatial-temporal graph modeling. By developing a novel adaptive dependency matrix and learn it through node embedding, our model can precisely capture the hidden spatial dependency in the data. With a stacked dilated 1D convolution component whose receptive field grows exponentially as the number of layers increases, Graph WaveNet is able to handle very long sequences. These two components are integrated seamlessly in a unified framework and the whole framework is learned in an end-to-end manner. Experimental results on two public traffic network datasets, METR-LA and PEMS-BAY, demonstrate the superior performance of our algorithm.
Subfields: Artificial Intelligence, Building and Construction, Transportation
View Paper: https://www.semanticscholar.org/paper/fae129338c0899576524506008427f64477d3967
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Shirui Pan
- Philip S. Yu
- Chengqi Zhang
- Guodong Long
- Jing Jiang
- Charu C. Aggarwal
- Zezhi Shao
- Weiwei Jiang
In Building and Construction
In Transportation
Citations in the corpus
- FourierGNN: Rethinking Multivariate Time Series Forecasting from a Pure Graph Perspective (2023, positive): “For example, AGCRN [2] use a GCN [14] and a GRU [21], GraphWaveNet [17] use a GCN and a TCN [8], etc.”
- Spatial-Temporal Fusion Graph Neural Networks for Traffic Flow Forecasting (2020, methodology): “• GraphWaveNet: Graph WaveNet is a framework combines adaptive adjacency matrix into graph convolution with 1D dilated convolution(Wu et al. 2019).”
- Spatial-Temporal Transformer Networks for Traffic Flow Forecasting (2020, methodology): “[22] capture hidden spatial patterns beyond predefined graph topology through a learnable embedding for each nodes in graph, which largely improve the forecasting accuracy but their spatial dependencies are still fixed once trained.”
- Disentangling and Unifying Graph Convolutions for Skeleton-Based Action Recognition (2020, background): “Among different GNN variants, the Graph Convolutional Network (GCN) [17] was first introduced as a first-order approximation for localized spectral convolutions, but its simplicity as a mean neighborhood aggregator [49, 46] has quickly led many subsequent spatial GNN architectures [49, 1, 45, 7] and”
- Dynamic and Multi-faceted Spatio-temporal Deep Learning for Traffic Speed Forecasting (2021, methodology): “Except for our model, Graph Wavenet and MTGNN perform well generally which might benefit from that they adopt adaptive graph to model relationships between nodes, indicating that adaptive graph based methods could effectively exploit valuable but latent spatial dependencies from historical traffic d”
- Adaptive Graph Convolutional Recurrent Network for Traffic Forecasting (2020, methodology): “Comparing with the self-adaptive adjacent matrix in [5], DAGG module is simpler and the learned EA has better interpret-ability.”
- Connecting the Dots: Multivariate Time Series Forecasting with Graph Neural Networks (2020, methodology): “• Graph WaveNet: A spatial-temporal graph convolutional network, which integrates diffusion graph convolutions with 1D dilated convolutions [21].”
- Skeleton-Based Action Recognition With Multi-Stream Adaptive Graph Convolutional Networks (2019, background): “[42] combine the pre-defined graphs and the learnable graph for graph convolutional layer.”
- Learning Dynamics and Heterogeneity of Spatial-Temporal Graph Data for Traffic Forecasting (2021, methodology): “CNN-based STGNNs [9], [10], [11], [12] utilize 1D CNN along the temporal dimension to capture temporal features, and use graph convolutions to capture spatial features.”
- Graph Neural Network for Traffic Forecasting: A Survey (2021, methodology): “For example, the road network is naturally a graph, with the road intersection or segment as the node and the road connection as the edge, and with the graph as the input, several GNN-based models have achieved the outstanding performance than previous approaches in tasks including road traffic flow”
- Spatial-Temporal Synchronous Graph Convolutional Networks: A New Framework for Spatial-Temporal Network Data Forecasting (2020, positive): “DCRNN, STGCN, ASTGCN, and Graph WaveNet use two module to model spatial dependencies and temporal correlations respectively.”
- DSTAGNN: Dynamic Spatial-Temporal Aware Graph Neural Network for Traffic Flow Forecasting (2022, positive): “Graph WaveNet (Wu et al., 2019) and AGCRN (Bai et al., 2020) found hidden spatial dependency through learnable embedding from nodes, but with these models, the spatial-temporal layers cannot be stacked while expanding the receptive field.”
- Decoupled Dynamic Spatial-Temporal Graph Neural Network for Traffic Forecasting (2022, methodology): “, knowledge-driven [3] and data-driven [7, 18, 21, 40].”
- ETA Prediction with Graph Neural Networks in Google Maps (2021, methodology): “This generic blueprint of a spatial GNN component combined with a temporal component yielded several subsequent proposals, including STGCN [34], GaAN [37], Graph WaveNet [29], ST-GRAT [19], StemGNN [3], and GMAN [39].”
- Graph Neural Networks for Intelligent Transportation Systems: A Survey (2023, methodology): “In comparison to ARIMA, SVR, FNN, FC-LSTM [28], STGCN [27], DCRNN [6], and Graph WaveNet [34], GMAN showed to perform better, especially in predicting the farther future.”
- Multi-Range Attentive Bicomponent Graph Convolutional Network for Traffic Forecasting (2019, methodology): “GCNs are building blocks for learning data with non-Euclidean structures, i.e., graphs (Wu et al., 2019b).”
- A Comprehensive Survey on Graph Neural Networks (2019, methodology): “To realize this, Graph WaveNet [76] proposes a self-adaptive adjacency matrix to perform graph convolutions.”