Graph neural networkFlow predictionSpatio-temporal characteristicThe COVID-19 pandemic highlighted significant challenges in maritime industries, such as port congestion and supply chain disruptions, necessitating accurate port traffic prediction. Traditional models often fail to capture the complex and dynamic...
//doi.org/10.1080/17538947.2023.2220610 REVIEW ARTICLE Advances in spatiotemporal graph neural network prediction research Yi Wang School of Earth and Space Sciences, Peking University, Beijing, People's Republic of China ABSTRACT Being a kind of non-Euclidean data, spatiotemporal graph data exists ...
spatio-temporal graphneural networkis proposed for predicting pedestrian trajectories. There are two components in the proposed method: spatial graphneural networkfor interaction modeling, and temporal graphneural networkfor motion feature extraction. Spatial graphneural networkuses an attention mechanism to ...
more advanced neighborhood aggregation methods54, scalable inference55and domain-specific applications20have been introduced. In general, a graph neural network performsRrounds of message passing, after which all nodes’ latent features
This study presents a spatio-temporal graph neural network (STGNN) model for streamflow prediction in the Upper Colorado River Basin (UCRB), integrating graph convolutional networks (GCNs) to model spatial connectivity and long short-term memory (LSTM) networks to capture temporal dynamics. Using ...
To improve the prediction accuracy of traffic flow under the influence of nearby time traffic flow disturbance, a dynamic spatiotemporal graph attention network traffic flow prediction model based on the attention mechanism was proposed. Considering the
Learning Heterogeneous Interaction Strengths by Trajectory Prediction with Graph Neural Network. arXiv preprint arXiv:2208.13179, 2022. Zhu Q (1991) Hidden Markov model for dynamic obstacle avoidance of mobile robot navigation. IEEE Trans Robot Autom 7(3):390–397 Article MathSciNet Google Scholar ...
ST-Meta Graph Reconstruction进一步设计用于通过重建不同城市的结构关系来进行结构感知元训练。 ST-GFSL 的端到端学习过程遵循基于MAML的episode learning。通过模拟目标城市的小样本场景,对批量的小样本训练任务进行采样,得到适应性强的基础模型。 Spatio-Temporal Neural Network ...
CROSS-NODE FEDERATED GRAPHNEURAL NETWORK 问题定义 给定一个图 G = (V, E) 的数据集,一个特征张量 X 和标签张量 Y,任务在数据集上定义,X 作为输入,Y 作为预测目标。我们考虑在跨节点联邦学习约束下学习模型:节点特征 ,…,节点标签 ,…,模型输出 ...
AGFCRN [74]: AGFCRN is an adaptive graph fusion convolutional recurrent network for traffic flow prediction. AST-InceptionNet [75]: AST-InceptionNet is a multi-scale adaptive spatio-temporal graph neural network model. DDGCRN [76]: DDGCRN is a Decomposition Dynamic Graph Convolutional Recurrent ...