Graph-based Forecasting with Missing Data through Spatiotemporal Downsampling

Ivan Marisca, Cesare Alippi, Filippo Maria Bianchi
Proceedings of the 41st International Conference on Machine Learning, PMLR 235:34846-34865, 2024.

Abstract

Given a set of synchronous time series, each associated with a sensor-point in space and characterized by inter-series relationships, the problem of spatiotemporal forecasting consists of predicting future observations for each point. Spatiotemporal graph neural networks achieve striking results by representing the relationships across time series as a graph. Nonetheless, most existing methods rely on the often unrealistic assumption that inputs are always available and fail to capture hidden spatiotemporal dynamics when part of the data is missing. In this work, we tackle this problem through hierarchical spatiotemporal downsampling. The input time series are progressively coarsened over time and space, obtaining a pool of representations that capture heterogeneous temporal and spatial dynamics. Conditioned on observations and missing data patterns, such representations are combined by an interpretable attention mechanism to generate the forecasts. Our approach outperforms state-of-the-art methods on synthetic and real-world benchmarks under different missing data distributions, particularly in the presence of contiguous blocks of missing values.

Cite this Paper

BibTeX
@InProceedings{pmlr-v235-marisca24a, title = {Graph-based Forecasting with Missing Data through Spatiotemporal Downsampling}, author = {Marisca, Ivan and Alippi, Cesare and Bianchi, Filippo Maria}, booktitle = {Proceedings of the 41st International Conference on Machine Learning}, pages = {34846--34865}, year = {2024}, editor = {Salakhutdinov, Ruslan and Kolter, Zico and Heller, Katherine and Weller, Adrian and Oliver, Nuria and Scarlett, Jonathan and Berkenkamp, Felix}, volume = {235}, series = {Proceedings of Machine Learning Research}, month = {21--27 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v235/main/assets/marisca24a/marisca24a.pdf}, url = {https://proceedings.mlr.press/v235/marisca24a.html}, abstract = {Given a set of synchronous time series, each associated with a sensor-point in space and characterized by inter-series relationships, the problem of spatiotemporal forecasting consists of predicting future observations for each point. Spatiotemporal graph neural networks achieve striking results by representing the relationships across time series as a graph. Nonetheless, most existing methods rely on the often unrealistic assumption that inputs are always available and fail to capture hidden spatiotemporal dynamics when part of the data is missing. In this work, we tackle this problem through hierarchical spatiotemporal downsampling. The input time series are progressively coarsened over time and space, obtaining a pool of representations that capture heterogeneous temporal and spatial dynamics. Conditioned on observations and missing data patterns, such representations are combined by an interpretable attention mechanism to generate the forecasts. Our approach outperforms state-of-the-art methods on synthetic and real-world benchmarks under different missing data distributions, particularly in the presence of contiguous blocks of missing values.} }
Endnote
%0 Conference Paper %T Graph-based Forecasting with Missing Data through Spatiotemporal Downsampling %A Ivan Marisca %A Cesare Alippi %A Filippo Maria Bianchi %B Proceedings of the 41st International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2024 %E Ruslan Salakhutdinov %E Zico Kolter %E Katherine Heller %E Adrian Weller %E Nuria Oliver %E Jonathan Scarlett %E Felix Berkenkamp %F pmlr-v235-marisca24a %I PMLR %P 34846--34865 %U https://proceedings.mlr.press/v235/marisca24a.html %V 235 %X Given a set of synchronous time series, each associated with a sensor-point in space and characterized by inter-series relationships, the problem of spatiotemporal forecasting consists of predicting future observations for each point. Spatiotemporal graph neural networks achieve striking results by representing the relationships across time series as a graph. Nonetheless, most existing methods rely on the often unrealistic assumption that inputs are always available and fail to capture hidden spatiotemporal dynamics when part of the data is missing. In this work, we tackle this problem through hierarchical spatiotemporal downsampling. The input time series are progressively coarsened over time and space, obtaining a pool of representations that capture heterogeneous temporal and spatial dynamics. Conditioned on observations and missing data patterns, such representations are combined by an interpretable attention mechanism to generate the forecasts. Our approach outperforms state-of-the-art methods on synthetic and real-world benchmarks under different missing data distributions, particularly in the presence of contiguous blocks of missing values.
APA
Marisca, I., Alippi, C. & Bianchi, F.M.. (2024). Graph-based Forecasting with Missing Data through Spatiotemporal Downsampling. Proceedings of the 41st International Conference on Machine Learning, in Proceedings of Machine Learning Research 235:34846-34865 Available from https://proceedings.mlr.press/v235/marisca24a.html.

Related Material