arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2511.18539cs.LGcs.CV
收藏

TimePre: 联合精度、效率与稳定性在概率时间序列预测中的应用

TimePre: Bridging Accuracy, Efficiency, and Stability in Probabilistic Time-Series Forecasting

Tohoku University(东京大学);WPI;Texas A&M University(德克萨斯大学);University of Wisconsin-Madison(威斯康星大学麦迪逊分校);UIUC(伊利诺伊大学香槟分校);Georgia Tech(佐治亚理工学院);MIT(麻省理工学院);University of Southern California(南加州大学);San Diego State University(圣地亚哥州立大学)
Lingyu Jiang, Lingyu Xu, Peiran Li, Dengzhe Hou, Qianwen Ge, Dingyi Zhuang, Shuo Xing, Wenjing Chen, Xiangbo Gao, Ting-Hsuan Chen, Xueying Zhan, Xin Zhang, Zimi… 展开作者

Lingyu Jiang, Lingyu Xu, Peiran Li, Dengzhe Hou, Qianwen Ge, Dingyi Zhuang, Shuo Xing, Wenjing Chen, Xiangbo Gao, Ting-Hsuan Chen, Xueying Zhan, Xin Zhang, Ziming Zhang, Zhengzhong Tu, Michael Zielewski, Kazunori Yamada, Fangzhou Lin

AI总结:

TimePre结合MLP模型的高效性与MCL方法的分布灵活性,通过SIN层解决精度、效率与稳定性之间的平衡问题,实现在概率时间序列预测中的高精度与高速度。

摘要:

我们提出TimePre,一个简单框架,将MLP模型的高效性与MCL方法的分布灵活性结合,用于概率时间序列预测。SIN层是TimePre的核心,通过修正通道统计偏移来稳定混合架构,解决灾难性假设崩溃问题。在六个基准数据集上的广泛实验表明,TimePre在关键概率度量上实现了最先进的准确性。关键的是,TimePre的推理速度比基于采样的模型快多个数量级,并且比先前的MCL方法更稳定。

英文摘要:

We propose TimePre, a simple framework that unifies the efficiency of Multilayer Perceptron (MLP)-based models with the distributional flexibility of Multiple Choice Learning (MCL) for Probabilistic Time-Series Forecasting (PTSF). Stabilized Instance Normalization (SIN), the core of TimePre, is a normalization layer that explicitly addresses the trade-off among accuracy, efficiency, and stability. SIN stabilizes the hybrid architecture by correcting channel-wise statistical shifts, thereby resolving the catastrophic hypothesis collapse. Extensive experiments on six benchmark datasets demonstrate that TimePre achieves state-of-the-art (SOTA) accuracy on key probabilistic metrics. Critically, TimePre achieves inference speeds that are orders of magnitude faster than sampling-based models, and is more stable than prior MCL approaches.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑