Irregular Multivariate Time Series Forecasting: A Transformable Patching Graph Neural Networks Approach

Weijia Zhang, Chenlong Yin, Hao Liu, Xiaofang Zhou, Hui Xiong
Proceedings of the 41st International Conference on Machine Learning, PMLR 235:60179-60196, 2024.

Abstract

Forecasting of Irregular Multivariate Time Series (IMTS) is critical for numerous areas, such as healthcare, biomechanics, climate science, and astronomy. Despite existing research addressing irregularities in time series through ordinary differential equations, the challenge of modeling correlations between asynchronous IMTS remains underexplored. To bridge this gap, this study proposes Transformable Patching Graph Neural Networks (t-PatchGNN), which transforms each univariate irregular time series into a series of transformable patches encompassing a varying number of observations with uniform temporal resolution. It seamlessly facilitates local semantics capture and inter-time series correlation modeling while avoiding sequence length explosion in aligned IMTS. Building on the aligned patching outcomes, we then present time-adaptive graph neural networks to model dynamic intertime series correlation based on a series of learned time-varying adaptive graphs. We demonstrate the remarkable superiority of t-PatchGNN on a comprehensive IMTS forecasting benchmark we build, which contains four real-world scientific datasets covering healthcare, biomechanics and climate science, and seventeen competitive baselines adapted from relevant research fields.

Cite this Paper

BibTeX
@InProceedings{pmlr-v235-zhang24bw, title = {Irregular Multivariate Time Series Forecasting: A Transformable Patching Graph Neural Networks Approach}, author = {Zhang, Weijia and Yin, Chenlong and Liu, Hao and Zhou, Xiaofang and Xiong, Hui}, booktitle = {Proceedings of the 41st International Conference on Machine Learning}, pages = {60179--60196}, 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/zhang24bw/zhang24bw.pdf}, url = {https://proceedings.mlr.press/v235/zhang24bw.html}, abstract = {Forecasting of Irregular Multivariate Time Series (IMTS) is critical for numerous areas, such as healthcare, biomechanics, climate science, and astronomy. Despite existing research addressing irregularities in time series through ordinary differential equations, the challenge of modeling correlations between asynchronous IMTS remains underexplored. To bridge this gap, this study proposes Transformable Patching Graph Neural Networks (t-PatchGNN), which transforms each univariate irregular time series into a series of transformable patches encompassing a varying number of observations with uniform temporal resolution. It seamlessly facilitates local semantics capture and inter-time series correlation modeling while avoiding sequence length explosion in aligned IMTS. Building on the aligned patching outcomes, we then present time-adaptive graph neural networks to model dynamic intertime series correlation based on a series of learned time-varying adaptive graphs. We demonstrate the remarkable superiority of t-PatchGNN on a comprehensive IMTS forecasting benchmark we build, which contains four real-world scientific datasets covering healthcare, biomechanics and climate science, and seventeen competitive baselines adapted from relevant research fields.} }
Endnote
%0 Conference Paper %T Irregular Multivariate Time Series Forecasting: A Transformable Patching Graph Neural Networks Approach %A Weijia Zhang %A Chenlong Yin %A Hao Liu %A Xiaofang Zhou %A Hui Xiong %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-zhang24bw %I PMLR %P 60179--60196 %U https://proceedings.mlr.press/v235/zhang24bw.html %V 235 %X Forecasting of Irregular Multivariate Time Series (IMTS) is critical for numerous areas, such as healthcare, biomechanics, climate science, and astronomy. Despite existing research addressing irregularities in time series through ordinary differential equations, the challenge of modeling correlations between asynchronous IMTS remains underexplored. To bridge this gap, this study proposes Transformable Patching Graph Neural Networks (t-PatchGNN), which transforms each univariate irregular time series into a series of transformable patches encompassing a varying number of observations with uniform temporal resolution. It seamlessly facilitates local semantics capture and inter-time series correlation modeling while avoiding sequence length explosion in aligned IMTS. Building on the aligned patching outcomes, we then present time-adaptive graph neural networks to model dynamic intertime series correlation based on a series of learned time-varying adaptive graphs. We demonstrate the remarkable superiority of t-PatchGNN on a comprehensive IMTS forecasting benchmark we build, which contains four real-world scientific datasets covering healthcare, biomechanics and climate science, and seventeen competitive baselines adapted from relevant research fields.
APA
Zhang, W., Yin, C., Liu, H., Zhou, X. & Xiong, H.. (2024). Irregular Multivariate Time Series Forecasting: A Transformable Patching Graph Neural Networks Approach. Proceedings of the 41st International Conference on Machine Learning, in Proceedings of Machine Learning Research 235:60179-60196 Available from https://proceedings.mlr.press/v235/zhang24bw.html.

Related Material