MSCGrapher: Learning Multi-Scale Dynamic Correlations for Multivariate Time Series Forecasting

Xian Yang, Zhenguo Zhang, Shihao Lu
Proceedings of the Forty-first Conference on Uncertainty in Artificial Intelligence, PMLR 286:4731-4751, 2025.

Abstract

Efficient learning intra-series and inter-series correlations is essential for multivariate time series forecasting (MTSF). However, in real-world scenarios, persistent and significant inter-series correlations are challenging to be represented in a static way and the strength of correlations varies across different time scales. In this paper, we address this challenge by modeling the complex inter-series relationships through dynamical correlations, considering the varying strengths of correlations. We propose a novel MTSF model: MSCGrapher, which leverages an adaptive correlation learning block to uncover inter-series correlations across different scales. Concretely, time series are first decomposed into different scales based on their periodicities. The graph representation of MTS is then constructed and an adaptive correlation learning method is introduced to capture the inter-series correlations across different scales. To quantify the strength of these correlations, we compute correlation scores based on the characteristics of the graph edges and classify correlations as either $\textit{Strong}$ or $\textit{Weak}$. Finally, we employ a self-attention module to capture intra-series correlations and then fuse features from all scales to obtain the final representation. Extensive experiments on 12 real-world datasets show that MSCGrapher gains significant forecasting performance, highlighting the critical role of inter-series correlations in capturing implicit patterns for MTS.

Cite this Paper

BibTeX
@InProceedings{pmlr-v286-yang25c, title = {MSCGrapher: Learning Multi-Scale Dynamic Correlations for Multivariate Time Series Forecasting}, author = {Yang, Xian and Zhang, Zhenguo and Lu, Shihao}, booktitle = {Proceedings of the Forty-first Conference on Uncertainty in Artificial Intelligence}, pages = {4731--4751}, year = {2025}, editor = {Chiappa, Silvia and Magliacane, Sara}, volume = {286}, series = {Proceedings of Machine Learning Research}, month = {21--25 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v286/main/assets/yang25c/yang25c.pdf}, url = {https://proceedings.mlr.press/v286/yang25c.html}, abstract = {Efficient learning intra-series and inter-series correlations is essential for multivariate time series forecasting (MTSF). However, in real-world scenarios, persistent and significant inter-series correlations are challenging to be represented in a static way and the strength of correlations varies across different time scales. In this paper, we address this challenge by modeling the complex inter-series relationships through dynamical correlations, considering the varying strengths of correlations. We propose a novel MTSF model: MSCGrapher, which leverages an adaptive correlation learning block to uncover inter-series correlations across different scales. Concretely, time series are first decomposed into different scales based on their periodicities. The graph representation of MTS is then constructed and an adaptive correlation learning method is introduced to capture the inter-series correlations across different scales. To quantify the strength of these correlations, we compute correlation scores based on the characteristics of the graph edges and classify correlations as either $\textit{Strong}$ or $\textit{Weak}$. Finally, we employ a self-attention module to capture intra-series correlations and then fuse features from all scales to obtain the final representation. Extensive experiments on 12 real-world datasets show that MSCGrapher gains significant forecasting performance, highlighting the critical role of inter-series correlations in capturing implicit patterns for MTS.} }
Endnote
%0 Conference Paper %T MSCGrapher: Learning Multi-Scale Dynamic Correlations for Multivariate Time Series Forecasting %A Xian Yang %A Zhenguo Zhang %A Shihao Lu %B Proceedings of the Forty-first Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2025 %E Silvia Chiappa %E Sara Magliacane %F pmlr-v286-yang25c %I PMLR %P 4731--4751 %U https://proceedings.mlr.press/v286/yang25c.html %V 286 %X Efficient learning intra-series and inter-series correlations is essential for multivariate time series forecasting (MTSF). However, in real-world scenarios, persistent and significant inter-series correlations are challenging to be represented in a static way and the strength of correlations varies across different time scales. In this paper, we address this challenge by modeling the complex inter-series relationships through dynamical correlations, considering the varying strengths of correlations. We propose a novel MTSF model: MSCGrapher, which leverages an adaptive correlation learning block to uncover inter-series correlations across different scales. Concretely, time series are first decomposed into different scales based on their periodicities. The graph representation of MTS is then constructed and an adaptive correlation learning method is introduced to capture the inter-series correlations across different scales. To quantify the strength of these correlations, we compute correlation scores based on the characteristics of the graph edges and classify correlations as either $\textit{Strong}$ or $\textit{Weak}$. Finally, we employ a self-attention module to capture intra-series correlations and then fuse features from all scales to obtain the final representation. Extensive experiments on 12 real-world datasets show that MSCGrapher gains significant forecasting performance, highlighting the critical role of inter-series correlations in capturing implicit patterns for MTS.
APA
Yang, X., Zhang, Z. & Lu, S.. (2025). MSCGrapher: Learning Multi-Scale Dynamic Correlations for Multivariate Time Series Forecasting. Proceedings of the Forty-first Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research 286:4731-4751 Available from https://proceedings.mlr.press/v286/yang25c.html.

Related Material