Towards Reliable Time Series Forecasting under Future Uncertainty: Ambiguity and Novelty Rejection Mechanisms
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_ | 1866915213160218624 |
|---|---|
| author | Feng, Ninghui Lai, Songning Zhou, Xin Yang, Jiayu Feng, Kunlong Yin, Zhenxiao Zhou, Fobao Hu, Zhangyi Yue, Yutao Liang, Yuxuan Wang, Boyu Zhao, Hang |
| author_facet | Feng, Ninghui Lai, Songning Zhou, Xin Yang, Jiayu Feng, Kunlong Yin, Zhenxiao Zhou, Fobao Hu, Zhangyi Yue, Yutao Liang, Yuxuan Wang, Boyu Zhao, Hang |
| contents | In real-world time series forecasting, uncertainty and lack of reliable evaluation pose significant challenges. Notably, forecasting errors often arise from underfitting in-distribution data and failing to handle out-of-distribution inputs. To enhance model reliability, we introduce a dual rejection mechanism combining ambiguity and novelty rejection. Ambiguity rejection, using prediction error variance, allows the model to abstain under low confidence, assessed through historical error variance analysis without future ground truth. Novelty rejection, employing Variational Autoencoders and Mahalanobis distance, detects deviations from training data. This dual approach improves forecasting reliability in dynamic environments by reducing errors and adapting to data changes, advancing reliability in complex scenarios. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_19656 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Towards Reliable Time Series Forecasting under Future Uncertainty: Ambiguity and Novelty Rejection Mechanisms Feng, Ninghui Lai, Songning Zhou, Xin Yang, Jiayu Feng, Kunlong Yin, Zhenxiao Zhou, Fobao Hu, Zhangyi Yue, Yutao Liang, Yuxuan Wang, Boyu Zhao, Hang Machine Learning Artificial Intelligence In real-world time series forecasting, uncertainty and lack of reliable evaluation pose significant challenges. Notably, forecasting errors often arise from underfitting in-distribution data and failing to handle out-of-distribution inputs. To enhance model reliability, we introduce a dual rejection mechanism combining ambiguity and novelty rejection. Ambiguity rejection, using prediction error variance, allows the model to abstain under low confidence, assessed through historical error variance analysis without future ground truth. Novelty rejection, employing Variational Autoencoders and Mahalanobis distance, detects deviations from training data. This dual approach improves forecasting reliability in dynamic environments by reducing errors and adapting to data changes, advancing reliability in complex scenarios. |
| title | Towards Reliable Time Series Forecasting under Future Uncertainty: Ambiguity and Novelty Rejection Mechanisms |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2503.19656 |