Spatiotemporal Prediction

Spatiotemporal prediction focuses on forecasting future states of systems that evolve over both space and time, aiming to improve accuracy and robustness across diverse applications. Current research emphasizes incorporating domain knowledge (e.g., physical laws, geographical features) into deep learning models, such as graph neural networks, recurrent neural networks (RNNs), and transformers, often coupled with techniques like uncertainty quantification and adaptive learning strategies to handle sparse or irregular data. These advancements are significantly impacting fields ranging from weather forecasting and traffic management to public health and environmental monitoring by enabling more accurate and reliable predictions in complex, real-world scenarios. The development of scalable and interpretable models remains a key focus.

Papers

August 14, 2022

DisenHCN: Disentangled Hypergraph Convolutional Networks for Spatiotemporal Activity Prediction
Yinfeng Li, Chen Gao, Quanming Yao, Tong Li, Depeng Jin, Yong Li
Spatio Temporal Heterogeneous Graph Hypergraph Neural Network Spatiotemporal Prediction Hypergraph Convolutional Network

June 24, 2022

Temporal Attention Unit: Towards Efficient Spatiotemporal Predictive Learning
Cheng Tan, Zhangyang Gao, Lirong Wu, Yongjie Xu, Jun Xia, Siyuan Li, Stan Z. Li
Temporal Attention Frame Attention Spatiotemporal Prediction Temporal Attention Module Spatiotemporal Predictive Learning

May 8, 2022

Adaptive Graph Convolutional Network Framework for Multidimensional Time Series Prediction
Ning Wang
Multivariate Time Series Forecasting Time Series Prediction Adaptive Graph Spatiotemporal Prediction Multivariate Forecasting

January 10, 2022

GridTuner: Reinvestigate Grid Size Selection for Spatiotemporal Prediction Models [Technical Report]
Jiabao Jin, Peng Cheng, Lei Chen, Xuemin Lin, Wenjie Zhang
Technical Report Traffic Prediction Spatiotemporal Prediction Grid Based Grid Search

December 10, 2021

Building Autocorrelation-Aware Representations for Fine-Scale Spatiotemporal Prediction
Yijun Lin, Yao-Yi Chiang, Meredith Franklin, Sandrah P. Eckel, José Luis Ambite
Spatiotemporal Representation SpatioTemporal Data Building PCC Spatiotemporal Prediction Air Quality Forecasting Spatiotemporal Learning Representation Matrix Time sEries

Spatiotemporal Prediction

Papers

DisenHCN: Disentangled Hypergraph Convolutional Networks for Spatiotemporal Activity Prediction

Temporal Attention Unit: Towards Efficient Spatiotemporal Predictive Learning

Adaptive Graph Convolutional Network Framework for Multidimensional Time Series Prediction

GridTuner: Reinvestigate Grid Size Selection for Spatiotemporal Prediction Models [Technical Report]

Building Autocorrelation-Aware Representations for Fine-Scale Spatiotemporal Prediction