Time-o1: Time-Series Forecasting Needs Transformed Label Alignment
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866912973234110464 |
|---|---|
| author | Wang, Hao Pan, Licheng Chen, Zhichao Chen, Xu Dai, Qingyang Wang, Lei Li, Haoxuan Lin, Zhouchen |
| author_facet | Wang, Hao Pan, Licheng Chen, Zhichao Chen, Xu Dai, Qingyang Wang, Lei Li, Haoxuan Lin, Zhouchen |
| contents | Training time-series forecast models presents unique challenges in designing effective learning objectives. Existing methods predominantly utilize the temporal mean squared error, which faces two critical challenges: (1) label autocorrelation, which leads to bias from the label sequence likelihood; (2) excessive amount of tasks, which increases with the forecast horizon and complicates optimization. To address these challenges, we propose Time-o1, a transformation-augmented learning objective tailored for time-series forecasting. The central idea is to transform the label sequence into decorrelated components with discriminated significance. Models are then trained to align the most significant components, thereby effectively mitigating label autocorrelation and reducing task amount. Extensive experiments demonstrate that Time-o1 achieves state-of-the-art performance and is compatible with various forecast models. Code is available at https://github.com/Master-PLC/Time-o1. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_17847 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Time-o1: Time-Series Forecasting Needs Transformed Label Alignment Wang, Hao Pan, Licheng Chen, Zhichao Chen, Xu Dai, Qingyang Wang, Lei Li, Haoxuan Lin, Zhouchen Machine Learning Artificial Intelligence Systems and Control Training time-series forecast models presents unique challenges in designing effective learning objectives. Existing methods predominantly utilize the temporal mean squared error, which faces two critical challenges: (1) label autocorrelation, which leads to bias from the label sequence likelihood; (2) excessive amount of tasks, which increases with the forecast horizon and complicates optimization. To address these challenges, we propose Time-o1, a transformation-augmented learning objective tailored for time-series forecasting. The central idea is to transform the label sequence into decorrelated components with discriminated significance. Models are then trained to align the most significant components, thereby effectively mitigating label autocorrelation and reducing task amount. Extensive experiments demonstrate that Time-o1 achieves state-of-the-art performance and is compatible with various forecast models. Code is available at https://github.com/Master-PLC/Time-o1. |
| title | Time-o1: Time-Series Forecasting Needs Transformed Label Alignment |
| topic | Machine Learning Artificial Intelligence Systems and Control |
| url | https://arxiv.org/abs/2505.17847 |