TimeBase: The Power of Minimalism in Efficient Long-term Time Series Forecasting
文献类型：期刊论文
| 作者 | Huang, Qihe1; Zhou, Zhengyang1,2,3; Yang, Kuo1; Yi, Zhongchao1; Wang, Xu1,3; Wang, Yang1,3 |
| 刊名 | INTERNATIONAL CONFERENCE ON MACHINE LEARNING
 |
| 出版日期 | 2025 |
| 卷号 | 267页码:26227-26246 |
| ISSN号 | 2640-3498 |
| 产权排序 | 3 |
| 文献子类 | Proceedings Paper |
| 英文摘要 | Long-term time series forecasting (LTSF) has traditionally relied on large parameters to capture extended temporal dependencies, resulting in substantial computational costs and inefficiencies in both memory usage and processing time. However, time series data, unlike high-dimensional images or text, often exhibit temporal pattern similarity and low-rank structures, especially in long-term horizons. By leveraging this structure, models can be guided to focus on more essential, concise temporal data, improving both accuracy and computational efficiency. In this paper, we introduce TimeBase, an ultra-lightweight network to harness the power of minimalism in LTSF. TimeBase 1) extracts core basis temporal components and 2) transforms traditional point-level forecasting into efficient segment-level forecasting, achieving optimal utilization of both data and parameters. Extensive experiments on diverse real-world datasets show that TimeBase achieves remarkable efficiency and secures competitive forecasting performance. Additionally, TimeBase can also serve as a very effective plug-and-play complexity reducer for any patch-based forecasting models. Code is available at https://github.com/hqh0728/TimeBase. |
| WOS研究方向 | Computer Science |
| 语种 | 英语 |
| WOS记录号 | WOS:001693126000131 |
| 出版者 | JMLR-JOURNAL MACHINE LEARNING RESEARCH |
| 源URL | [http://ir.igsnrr.ac.cn/handle/311030/221371]  |
| 专题 | 资源与环境信息系统国家重点实验室_外文论文 |
| 通讯作者 | Zhou, Zhengyang; Wang, Yang |
| 作者单位 | 1.Univ Sci & Technol China USTC, Hefei, Peoples R China; 2.State Key Lab Resources & Environm Informat Syst, Beijing, Peoples R China 3.USTC, Suzhou Inst Adv Res, Suzhou, Peoples R China; |
| 推荐引用方式 GB/T 7714 | Huang, Qihe,Zhou, Zhengyang,Yang, Kuo,et al. TimeBase: The Power of Minimalism in Efficient Long-term Time Series Forecasting[J]. INTERNATIONAL CONFERENCE ON MACHINE LEARNING,2025,267:26227-26246. |
| APA | Huang, Qihe,Zhou, Zhengyang,Yang, Kuo,Yi, Zhongchao,Wang, Xu,&Wang, Yang.(2025).TimeBase: The Power of Minimalism in Efficient Long-term Time Series Forecasting.INTERNATIONAL CONFERENCE ON MACHINE LEARNING,267,26227-26246. |
| MLA | Huang, Qihe,et al."TimeBase: The Power of Minimalism in Efficient Long-term Time Series Forecasting".INTERNATIONAL CONFERENCE ON MACHINE LEARNING 267(2025):26227-26246. |
入库方式： OAI收割
来源：地理科学与资源研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。