Causal-oriented representation learning for time-series forecasting based on the spatiotemporal information transformation
Abstract In real-world high-dimensional systems, both causal dependencies and temporal information of key variables are essential for dissecting the underlying mechanisms governing system dynamics. However, effective approaches to synthesize these two interconnected aspects for deeper insights remai...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-06-01
|
| Series: | Communications Physics |
| Online Access: | https://doi.org/10.1038/s42005-025-02170-6 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849691374295711744 |
|---|---|
| author | Sihua Cai Hao Peng Rui Liu Pei Chen |
| author_facet | Sihua Cai Hao Peng Rui Liu Pei Chen |
| author_sort | Sihua Cai |
| collection | DOAJ |
| description | Abstract In real-world high-dimensional systems, both causal dependencies and temporal information of key variables are essential for dissecting the underlying mechanisms governing system dynamics. However, effective approaches to synthesize these two interconnected aspects for deeper insights remain lacking. Here we show a neural network framework, the Causal-oriented Representation Learning Predictor (CReP), which jointly conducts causal analysis and multistep forecasting from a unified perspective. CReP implicitly learns latent causal representations from observed data while simultaneously making multistep predictions, and explicitly interprets the representations to uncover the causes and effects of target variable. The core idea of CReP is to decompose the original space into three orthogonal latent factors, each capturing distinct causal representations: cause-related, effect-related, and non-causal representations of the target variable. The reconstruction-based dynamic causation, generalized through spatiotemporal information (STI) transformation mechanism, provides a theoretical foundation for simultaneously modeling causal interactions via latent representations and predicting future states using the effect representation. Evaluations on three simulation models and two real-world datasets demonstrate CReP’s robust forecasting accuracy and reliable causal insights. As a self-supervised-learning approach, CReP shows significant potential for practical applications and provides a unified framework to reveal intrinsic mechanisms in dynamical systems by integrating causal relationships and temporal information. |
| format | Article |
| id | doaj-art-f07de10953e140c5a369ff7dd954ea46 |
| institution | DOAJ |
| issn | 2399-3650 |
| language | English |
| publishDate | 2025-06-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Communications Physics |
| spelling | doaj-art-f07de10953e140c5a369ff7dd954ea462025-08-20T03:21:03ZengNature PortfolioCommunications Physics2399-36502025-06-018111410.1038/s42005-025-02170-6Causal-oriented representation learning for time-series forecasting based on the spatiotemporal information transformationSihua Cai0Hao Peng1Rui Liu2Pei Chen3School of Mathematics, South China University of TechnologySchool of Mathematics, South China University of TechnologySchool of Mathematics, South China University of TechnologySchool of Mathematics, South China University of TechnologyAbstract In real-world high-dimensional systems, both causal dependencies and temporal information of key variables are essential for dissecting the underlying mechanisms governing system dynamics. However, effective approaches to synthesize these two interconnected aspects for deeper insights remain lacking. Here we show a neural network framework, the Causal-oriented Representation Learning Predictor (CReP), which jointly conducts causal analysis and multistep forecasting from a unified perspective. CReP implicitly learns latent causal representations from observed data while simultaneously making multistep predictions, and explicitly interprets the representations to uncover the causes and effects of target variable. The core idea of CReP is to decompose the original space into three orthogonal latent factors, each capturing distinct causal representations: cause-related, effect-related, and non-causal representations of the target variable. The reconstruction-based dynamic causation, generalized through spatiotemporal information (STI) transformation mechanism, provides a theoretical foundation for simultaneously modeling causal interactions via latent representations and predicting future states using the effect representation. Evaluations on three simulation models and two real-world datasets demonstrate CReP’s robust forecasting accuracy and reliable causal insights. As a self-supervised-learning approach, CReP shows significant potential for practical applications and provides a unified framework to reveal intrinsic mechanisms in dynamical systems by integrating causal relationships and temporal information.https://doi.org/10.1038/s42005-025-02170-6 |
| spellingShingle | Sihua Cai Hao Peng Rui Liu Pei Chen Causal-oriented representation learning for time-series forecasting based on the spatiotemporal information transformation Communications Physics |
| title | Causal-oriented representation learning for time-series forecasting based on the spatiotemporal information transformation |
| title_full | Causal-oriented representation learning for time-series forecasting based on the spatiotemporal information transformation |
| title_fullStr | Causal-oriented representation learning for time-series forecasting based on the spatiotemporal information transformation |
| title_full_unstemmed | Causal-oriented representation learning for time-series forecasting based on the spatiotemporal information transformation |
| title_short | Causal-oriented representation learning for time-series forecasting based on the spatiotemporal information transformation |
| title_sort | causal oriented representation learning for time series forecasting based on the spatiotemporal information transformation |
| url | https://doi.org/10.1038/s42005-025-02170-6 |
| work_keys_str_mv | AT sihuacai causalorientedrepresentationlearningfortimeseriesforecastingbasedonthespatiotemporalinformationtransformation AT haopeng causalorientedrepresentationlearningfortimeseriesforecastingbasedonthespatiotemporalinformationtransformation AT ruiliu causalorientedrepresentationlearningfortimeseriesforecastingbasedonthespatiotemporalinformationtransformation AT peichen causalorientedrepresentationlearningfortimeseriesforecastingbasedonthespatiotemporalinformationtransformation |