# Spatiotemporal Attention for Multivariate Time Series Prediction and Interpretation

arXiv: https://arxiv.org/abs/2008.04882 · HTML: https://mlsanity.com/paper/2008.04882/

## Key facts

- Cited by 18 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 11 Aug 2020.

## Built on by (most cited first, 18 of 18)

- [Long-Range Transformers for Dynamic Spatiotemporal Forecasting](https://mlsanity.com/paper/2109.12218/) (2021)
- [Interpretation of Time-Series Deep Models: A Survey](https://mlsanity.com/paper/2305.14582/) (2023)
- [Solving Dynamic Graph Problems with Multi-Attention Deep Reinforcement Learning](https://mlsanity.com/paper/2201.04895/) (2022)
- [TSEM: Temporally Weighted Spatiotemporal Explainable Neural Network for Multivariate Time Series](https://mlsanity.com/paper/2205.13012/) (2022)
- [Self-explaining Hierarchical Model for Intraoperative Time Series](https://mlsanity.com/paper/2210.04417/) (2022)
- [Structured Temporal Causality for Interpretable Multivariate Time Series Anomaly Detection](https://mlsanity.com/paper/2510.16511/) (2025)
- [Earthquake Nowcasting with Deep Learning](https://mlsanity.com/paper/2201.01869/) (2021)
- [A Dynamic Systems Approach to Modelling Human-Machine Rhythm Interaction](https://mlsanity.com/paper/2407.09538/) (2024)
- [Cross-Modal Virtual Sensing for Combustion Instability Monitoring](https://mlsanity.com/paper/2110.01659/) (2021)
- [A Joint Learning Framework with Feature Reconstruction and Prediction for Incomplete Satellite Image Time Series in Agricultural Semantic Segmentation](https://mlsanity.com/paper/2505.19159/) (2025)
- [Fourier Basis Mapping: A Time-Frequency Learning Framework for Time Series Forecasting](https://mlsanity.com/paper/2507.09445/) (2025)
- [Evolutionary Ensemble Learning for Multivariate Time Series Prediction](https://mlsanity.com/paper/2108.09659/) (2021)
- [On the balance between the training time and interpretability of neural ODE for time series modelling](https://mlsanity.com/paper/2206.03304/) (2022)
- [Generative Learning for Simulation of Vehicle Faults](https://mlsanity.com/paper/2407.17654/) (2024)
- [Interpretable deep convolutional model for nonlinear multivariate time series in complex systems](https://mlsanity.com/paper/2501.04339/) (2025)
- [PHEATPRUNER: Interpretable Data-centric Feature Selection for Multivariate Time Series Classification through Persistent Homology](https://mlsanity.com/paper/2504.18329/) (2025)
- [TimeCatcher: A Variational Framework for Volatility-Aware Forecasting of Non-Stationary Time Series](https://mlsanity.com/paper/2601.20448/) (2026)
- [PatchDecomp: Interpretable Patch-Based Time Series Forecasting](https://mlsanity.com/paper/2603.03902/) (2026)

## Abstract (arXiv)

Multivariate time series modeling and prediction problems are abundant in many machine learning application domains. Accurate interpretation of such prediction outcomes from a machine learning model that explicitly captures temporal correlations can significantly benefit the domain experts. In this context, temporal attention has been successfully applied to isolate the important time steps for the input time series. However, in multivariate time series problems, spatial interpretation is also critical to understand the contributions of different variables on the model outputs. We propose a novel deep learning architecture, called spatiotemporal attention mechanism (STAM) for simultaneous learning of the most important time steps and variables. STAM is a causal (i.e., only depends on past inputs and does not use future inputs) and scalable (i.e., scales well with an increase in the number of variables) approach that is comparable to the state-of-the-art models in terms of computational tractability. We demonstrate our models' performance on two popular public datasets and a domain-specific dataset. When compared with the baseline models, the results show that STAM maintains state-of-the-art prediction accuracy while offering the benefit of accurate spatiotemporal interpretability. The learned attention weights are validated from a domain knowledge perspective for these real-world datasets.