Spatiotemporal Forecasting

Spatiotemporal forecasting aims to predict future values of data exhibiting both spatial and temporal dependencies, crucial for applications like weather prediction, traffic management, and energy forecasting. Current research emphasizes improving model efficiency and accuracy using deep learning architectures such as Graph Neural Networks (GNNs), Transformers, and recurrent networks (RNNs), often incorporating techniques like attention mechanisms, vector quantization, and knowledge distillation to handle high-dimensional data and complex relationships. These advancements are driving improvements in prediction accuracy and enabling the integration of diverse data sources, including numerical and visual data, leading to more reliable and informative forecasts across various domains. The resulting improvements have significant implications for resource management, risk mitigation, and decision-making in numerous sectors.

Papers

December 10, 2022

Scalable Dynamic Mixture Model with Full Covariance for Probabilistic Traffic Forecasting
Seongjin Choi, Nicolas Saunier, Vincent Zhihao Zheng, Martin Trepanier, Lijun Sun
Potential Scalability Traffic Forecasting Mixture Model Covariance Conditioning Spatiotemporal Forecasting Traffic Speed Long Term Traffic Forecasting

November 17, 2022

Sub-Graph Learning for Spatiotemporal Forecasting via Knowledge Distillation
Mehrtash Mehrabi, Yingxue Zhang
Knowledge Distillation Subgraph Structure Large Graph Local Model Spatiotemporal Modeling Spatiotemporal Forecasting Subgraph Level Training

November 7, 2022

Spatiotemporal forecasting of vertical track alignment with exogenous factors
Katsuya Kosukegawa, Yasukuni Mori, Hiroki Suyari, Kazuhiko Kawamoto
Spatiotemporal Forecasting Spatial Correlation Spatiotemporal Correlation Exogenous Variable Track Geometry

September 29, 2022

A case study of spatiotemporal forecasting techniques for weather forecasting
Shakir Showkat Sofi, Ivan Oseledets
Case Study Numerical Weather Prediction Weather Forecasting Climate Data Spatiotemporal Forecasting Spatiotemporal Prediction Spatiotemporal Correlation

August 17, 2022

Towards Learning in Grey Spatiotemporal Systems: A Prophet to Non-consecutive Spatiotemporal Dynamics
Zhengyang Zhou, Yang Kuo, Wei Sun, Binwu Wang, Min Zhou, Yunan Zong, Yang Wang
LeArning Abstract Multi Step Feature Decoupling Spatiotemporal Forecasting Spatiotemporal Learning Spatiotemporal Dynamic

August 2, 2022

A Novel Transformer Network with Shifted Window Cross-Attention for Spatiotemporal Weather Forecasting
Alabi Bojesomo, Hasan Al Marzouqi, Panos Liatsis
Swin Transformer Video Transformer Spatiotemporal Forecasting Vision Transformer Architecture Meteorological Satellite Weather4cast Challenge

June 21, 2022

Predicting Team Performance with Spatial Temporal Graph Convolutional Networks
Shengnan Hu, Gita Sukthankar
Deep Convolutional Neural Network Sport Analytics Spatiotemporal Forecasting Opponent Modeling Team Performance

January 21, 2022

A Systematic Exploration of Reservoir Computing for Forecasting Complex Spatiotemporal Dynamics
Jason A. Platt, Stephen G. Penny, Timothy A. Smith, Tse-Chun Chen, Henry D. I. Abarbanel
Reservoir Computing Active Exploration Chaotic System Spatiotemporal Forecasting Reservoir Computer Reservoir Model Chaotic Lorenz

January 17, 2022

SwinUNet3D -- A Hierarchical Architecture for Deep Traffic Prediction using Shifted Window Transformers
Alabi Bojesomo, Hasan Al Marzouqi, Panos Liatsis
Vision Transformer Swin Transformer Traffic Prediction Traffic Forecasting Spatiotemporal Forecasting Shifted Window Transformer

December 13, 2021

Efficient spatio-temporal weather forecasting using U-Net
Akshay Punjabi, Pablo Izquierdo Ayala
Supervised Autoencoder U Net Numerical Weather Prediction Spatiotemporal Forecasting Weather Prediction Task

Spatiotemporal Forecasting

Papers

Scalable Dynamic Mixture Model with Full Covariance for Probabilistic Traffic Forecasting

Sub-Graph Learning for Spatiotemporal Forecasting via Knowledge Distillation

Spatiotemporal forecasting of vertical track alignment with exogenous factors

A case study of spatiotemporal forecasting techniques for weather forecasting

Towards Learning in Grey Spatiotemporal Systems: A Prophet to Non-consecutive Spatiotemporal Dynamics

A Novel Transformer Network with Shifted Window Cross-Attention for Spatiotemporal Weather Forecasting

Predicting Team Performance with Spatial Temporal Graph Convolutional Networks

A Systematic Exploration of Reservoir Computing for Forecasting Complex Spatiotemporal Dynamics

SwinUNet3D -- A Hierarchical Architecture for Deep Traffic Prediction using Shifted Window Transformers

Efficient spatio-temporal weather forecasting using U-Net