APT: Affine Prototype-Timestamp For Time Series Forecasting Under Distribution Shift
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866915621754634240 |
|---|---|
| author | Li, Yujie Shao, Zezhi Yu, Chengqing Fu, Yisong Sun, Tao Xu, Yongjun Wang, Fei |
| author_facet | Li, Yujie Shao, Zezhi Yu, Chengqing Fu, Yisong Sun, Tao Xu, Yongjun Wang, Fei |
| contents | Time series forecasting under distribution shift remains challenging, as existing deep learning models often rely on local statistical normalization (e.g., mean and variance) that fails to capture global distribution shift. Methods like RevIN and its variants attempt to decouple distribution and pattern but still struggle with missing values, noisy observations, and invalid channel-wise affine transformation. To address these limitations, we propose Affine Prototype Timestamp (APT), a lightweight and flexible plug-in module that injects global distribution features into the normalization-forecasting pipeline. By leveraging timestamp conditioned prototype learning, APT dynamically generates affine parameters that modulate both input and output series, enabling the backbone to learn from self-supervised, distribution-aware clustered instances. APT is compatible with arbitrary forecasting backbones and normalization strategies while introducing minimal computational overhead. Extensive experiments across six benchmark datasets and multiple backbone-normalization combinations demonstrate that APT significantly improves forecasting performance under distribution shift. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_12945 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | APT: Affine Prototype-Timestamp For Time Series Forecasting Under Distribution Shift Li, Yujie Shao, Zezhi Yu, Chengqing Fu, Yisong Sun, Tao Xu, Yongjun Wang, Fei Machine Learning Time series forecasting under distribution shift remains challenging, as existing deep learning models often rely on local statistical normalization (e.g., mean and variance) that fails to capture global distribution shift. Methods like RevIN and its variants attempt to decouple distribution and pattern but still struggle with missing values, noisy observations, and invalid channel-wise affine transformation. To address these limitations, we propose Affine Prototype Timestamp (APT), a lightweight and flexible plug-in module that injects global distribution features into the normalization-forecasting pipeline. By leveraging timestamp conditioned prototype learning, APT dynamically generates affine parameters that modulate both input and output series, enabling the backbone to learn from self-supervised, distribution-aware clustered instances. APT is compatible with arbitrary forecasting backbones and normalization strategies while introducing minimal computational overhead. Extensive experiments across six benchmark datasets and multiple backbone-normalization combinations demonstrate that APT significantly improves forecasting performance under distribution shift. |
| title | APT: Affine Prototype-Timestamp For Time Series Forecasting Under Distribution Shift |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2511.12945 |