More Like this

1. Urban Traffic Dynamics Prediction—A Continuous Spatial-temporal Meta-learning Approach

   https://doi.org/10.1145/3474837  

   Zhang, Yingxue; Li, Yanhua; Zhou, Xun; Luo, Jun; Zhang, Zhi-Li (April 2022, ACM Transactions on Intelligent Systems and Technology)

   Urban traffic status (e.g., traffic speed and volume) is highly dynamic in nature, namely, varying across space and evolving over time. Thus, predicting such traffic dynamics is of great importance to urban development and transportation management. However, it is very challenging to solve this problem due to spatial-temporal dependencies and traffic uncertainties. In this article, we solve the traffic dynamics prediction problem from Bayesian meta-learning perspective and propose a novel continuous spatial-temporal meta-learner (cST-ML), which is trained on a distribution of traffic prediction tasks segmented by historical traffic data with the goal of learning a strategy that can be quickly adapted to related but unseen traffic prediction tasks. cST-ML tackles the traffic dynamics prediction challenges by advancing the Bayesian black-box meta-learning framework through the following new points: (1) cST-ML captures the dynamics of traffic prediction tasks using variational inference, and to better capture the temporal uncertainties within tasks, cST-ML performs as a rolling window within each task; (2) cST-ML has novel designs in architecture, where CNN and LSTM are embedded to capture the spatial-temporal dependencies between traffic status and traffic-related features; (3) novel training and testing algorithms for cST-ML are designed. We also conduct experiments on two real-world traffic datasets (taxi inflow and traffic speed) to evaluate our proposed cST-ML. The experimental results verify that cST-ML can significantly improve the urban traffic prediction performance and outperform all baseline models especially when obvious traffic dynamics and temporal uncertainties are presented. 

   more » « less
2. cST-ML: Continuous Spatial-Temporal Meta-Learning for Traffic Dynamics Prediction

   https://doi.org/10.1109/icdm50108.2020.00187  

   Zhang, Yingxue; Li, Yanhua; Zhou, Xun; Luo, Jun (November 2020, 2020 IEEE International Conference on Data Mining (ICDM))

   null (Ed.)

   Urban traffic status (e.g., traffic speed and volume) is highly dynamic in nature, namely, varying across space and evolving over time. Thus, predicting such traffic dynamics is of great importance to urban development and transportation management. However, it is very challenging to solve this problem due to spatial-temporal dependencies and traffic uncertainties. In this paper, we solve the traffic dynamics prediction problem from Bayesian meta-learning perspective and propose a novel continuous spatial-temporal meta-learner (cST-ML), which is trained on a distribution of traffic prediction tasks segmented by historical traffic data with the goal of learning a strategy that can be quickly adapted to related but unseen traffic prediction tasks. cST-ML tackles the traffic dynamics prediction challenges by advancing the Bayesian black-box meta-learning framework through the following new points: 1) cST-ML captures the dynamics of traffic prediction tasks using variational inference; 2) cST-ML has novel designs in architecture, where CNN and LSTM are embedded to capture the spatial-temporal dependencies between traffic status and traffic related features; 3) novel training and testing algorithms for cST-ML are designed. We also conduct experiments on two real-world traffic datasets (taxi inflow and traffic speed) to evaluate our proposed cST-ML. The experimental results verify that cST-ML can significantly improve the urban traffic prediction performance and outperform all baseline models. 

   more » « less
3. A cluster-based temporal attention approach for predicting cyclone-induced compound flood dynamics

   Daramola, Samuel; Muñoz, David F; Moftakhari, Hamed; Moradkhani, Hamid (June 2025, Environmental modelling software)

   Deep learning (DL) models have been used for rapid assessments of environmental phenomena like mapping compound flood hazards from cyclones. However, predicting compound flood dynamics (e.g., flood extent and inundation depth over time) is often done with physically-based models because they capture physical drivers, nonlinear interactions, and hysteresis in system behavior. Here, we show that a customized DL model can efficiently learn spatiotemporal dependencies of multiple flood events in Galveston, TX. The proposed model combines the spatial feature extraction of CNN, temporal regression of LSTM, and a novel cluster-based temporal attention approach to assimilate multimodal inputs; thus, accurately replicating compound flood dynamics of physically-based models. The DL model achieves satisfactory flood timing (±1 h), critical success index above 60 %, RMSE below 0.10 m, and nearly perfect error bias of 1. These results demonstrate the model's potential to assist in flood preparation and response efforts in vulnerable coastal regions. 

   more » « less
4. Comparison of CNN and LSTM Networks on Human Intention Prediction in Physical Human-Robot Interactions

   Zadeh, Khosro G; Zendehdel, Niloofar; Holmes, George L; Bonnett, Keyri M; Costa, Amy; Burns, Devin; Leu, Ming C; Song, Yun Seong (August 2025, The Fool)

   Advancements in robotics and AI have increased the demand for interactive robots in healthcare and assistive applications. However, ensuring safe and effective physical human-robot interactions (pHRIs) remains challenging due to the complexities of human motor communication and intent recognition. Traditional physics-based models struggle to capture the dynamic nature of human force interactions, limiting robotic adaptability. To address these limitations, neural networks (NNs) have been explored for force-movement intention prediction. While multi-layer perceptron (MLP) networks show potential, they struggle with temporal dependencies and generalization. Long Short-Term Memory (LSTM) networks effectively model sequential dependencies, while Convolutional Neural Networks (CNNs) enhance spatial feature extraction from human force data. Building on these strengths, this study introduces a hybrid LSTM-CNN framework to improve force-movement intention prediction, increasing accuracy from 69% to 86% through effective denoising and advanced architectures. The combined CNN-LSTM network proved particularly effective in handling individualized force-velocity relationships and presents a generalizable model paving the way for more adaptive strategies in robot guidance. These findings highlight the importance of integrating spatial and temporal modeling to enhance robot precision, responsiveness, and human-robot collaboration. Index Terms —- Physical Human-Robot Interaction, Intention Detection, Machine Learning, Long-Short Term Memory (LSTM) 

   more » « less
5. Multi-scale dynamic spatiotemporal graph attention network for forecasting karst spring discharge

   https://doi.org/10.1016/j.jhydrol.2025.133289  

   Zhou, Renjie (October 2025, Journal of Hydrology)

   Karst aquifers are important groundwater resources that supply drinking water for approximately 25 % of the world’s population. Their complex hydrogeological structures, dual-flow regimes, and highly heterogeneous flow pose significant challenges for accurate hydrodynamic modeling and sustainable management. Traditional modeling approaches often struggle to capture the intricate spatial dependencies and multi-scale temporal patterns inherent in karst systems, particularly the interactions between rapid conduit flow and slower matrix flow. This study proposes a novel multi-scale dynamic graph attention network integrated with long short-term memory model (GAT-LSTM) to innovatively learn and integrate spatial and temporal dependencies in karst systems for forecasting spring discharge. The model introduces several innovative components: (1) graph-based neural networks with dynamic edge-weighting mechanism are proposed to learn and update spatial dependencies based on both geographic distances and learned hydrological relationships, (2) a multi-head attention mechanism is adopted to capture different aspects of spatial relationships simultaneously, and (3) a hierarchical temporal architecture is incorporated to process hydrological temporal patterns at both monthly and seasonal scales with an adaptive fusion mechanism for final results. These features enable the proposed model to effectively account for the dual-flow dynamics in karst systems, where rapid conduit flow and slower matrix flow coexist. The newly proposed model is applied to the Barton Springs of the Edwards Aquifer in Texas. The results demonstrate that it can obtain more accurate and robust prediction performance across various time steps compared to traditional temporal and spatial deep learning approaches. Based on the multi-scale GAT-LSTM model, a comprehensive ablation analysis and permutation feature important are conducted to analyze the relative contribution of various input variables on the final prediction. These findings highlight the intricate nature of karst systems and demonstrate that effective spring discharge prediction requires comprehensive monitoring networks encompassing both primary recharge contributors and supplementary hydrological features that may serve as valuable indicators of system-wide conditions. 

   more » « less