Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | https://arxiv.org/abs/2403.08572 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866910365897457664 |
|---|---|
| author | Zhang, Kexuan Zou, Xiaobei Tang, Yang |
| author_facet | Zhang, Kexuan Zou, Xiaobei Tang, Yang |
| contents | Time series analysis is a vital task with broad applications in various domains. However, effectively capturing cross-dimension and cross-time dependencies in non-stationary time series poses significant challenges, particularly in the context of environmental factors. The spurious correlation induced by the environment confounds the causal relationships between cross-dimension and cross-time dependencies. In this paper, we introduce a novel framework called Caformer (\underline{\textbf{Ca}}usal Trans\underline{\textbf{former}}) for time series analysis from a causal perspective. Specifically, our framework comprises three components: Dynamic Learner, Environment Learner, and Dependency Learner. The Dynamic Learner unveils dynamic interactions among dimensions, the Environment Learner mitigates spurious correlations caused by environment with a back-door adjustment, and the Dependency Learner aims to infer robust interactions across both time and dimensions. Our Caformer demonstrates consistent state-of-the-art performance across five mainstream time series analysis tasks, including long- and short-term forecasting, imputation, classification, and anomaly detection, with proper interpretability. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2403_08572 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Caformer: Rethinking Time Series Analysis from Causal Perspective Zhang, Kexuan Zou, Xiaobei Tang, Yang Machine Learning Time series analysis is a vital task with broad applications in various domains. However, effectively capturing cross-dimension and cross-time dependencies in non-stationary time series poses significant challenges, particularly in the context of environmental factors. The spurious correlation induced by the environment confounds the causal relationships between cross-dimension and cross-time dependencies. In this paper, we introduce a novel framework called Caformer (\underline{\textbf{Ca}}usal Trans\underline{\textbf{former}}) for time series analysis from a causal perspective. Specifically, our framework comprises three components: Dynamic Learner, Environment Learner, and Dependency Learner. The Dynamic Learner unveils dynamic interactions among dimensions, the Environment Learner mitigates spurious correlations caused by environment with a back-door adjustment, and the Dependency Learner aims to infer robust interactions across both time and dimensions. Our Caformer demonstrates consistent state-of-the-art performance across five mainstream time series analysis tasks, including long- and short-term forecasting, imputation, classification, and anomaly detection, with proper interpretability. |
| title | Caformer: Rethinking Time Series Analysis from Causal Perspective |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2403.08572 |