arXiv 2602.225202026-02-26cs.LG

TEFL: 基于预测残差的多时间尺度滚动预测

TEFL: Prediction-Residual-Guided Rolling Forecasting for Multi-Horizon Time Series

Key Laboratory of Road and Traffic Engineering, Ministry of Education Tongji University, Shanghai China（交通工程教育部重点实验室，同济大学，上海中国）；The College of Computer Science, Tongji University, Shanghai China（计算机科学学院，同济大学，上海中国）；Research Center for Scientific Data Hub, Zhejiang Lab, Hangzhou, China（科学数据中心研究中心，浙江实验室，杭州中国）

Xiannan Huang, Shen Fang, Shuhan Qiu, Chengcheng Yu, Jiayuan Du, Chao Yang

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

TEFL通过整合历史预测残差提升多时间尺度时间序列预测的准确性和鲁棒性。

摘要：

时间序列预测在交通、能源和气象等领域中发挥着关键作用。尽管取得了成功，但现代深度预测模型通常被训练以最小化点预测损失，而没有利用滚动预测中包含的丰富信息——这些残差反映了持续的偏差、未建模的模式或演化的动态。我们提出了TEFL（时间误差反馈学习），这是一个统一的学习框架，它在训练和评估过程中明确地将这些历史残差纳入预测流程中。为了使深度多步预测在实践中可行，我们解决了三个关键挑战：（1）在滚动预测的不完全可观察性下选择可观测的多步残差；（2）通过轻量级低秩适配器整合这些残差，以保持效率并防止过拟合；（3）设计一个两阶段训练过程，共同优化基础预测器和误差模块。在10个真实世界数据集和5个基础架构上进行了广泛的实验，结果显示TEFL在准确性方面始终有所提高，平均将MAE降低5-10%。此外，它在突然变化和分布转移下表现出强大的鲁棒性，在具有挑战性的场景中，误差减少超过10%（最高达19.5%）。通过将基于残差的反馈直接嵌入学习过程中，TEFL为现代深度预测系统提供了一个简单、通用且有效的增强。

英文摘要：

Time series forecasting plays a critical role in domains such as transportation, energy, and meteorology. Despite their success, modern deep forecasting models are typically trained to minimize point-wise prediction loss without leveraging the rich information contained in past prediction residuals from rolling forecasts - residuals that reflect persistent biases, unmodeled patterns, or evolving dynamics. We propose TEFL (Temporal Error Feedback Learning), a unified learning framework that explicitly incorporates these historical residuals into the forecasting pipeline during both training and evaluation. To make this practical in deep multi-step settings, we address three key challenges: (1) selecting observable multi-step residuals under the partial observability of rolling forecasts, (2) integrating them through a lightweight low-rank adapter to preserve efficiency and prevent overfitting, and (3) designing a two-stage training procedure that jointly optimizes the base forecaster and error module. Extensive experiments across 10 real-world datasets and 5 backbone architectures show that TEFL consistently improves accuracy, reducing MAE by 5-10% on average. Moreover, it demonstrates strong robustness under abrupt changes and distribution shifts, with error reductions exceeding 10% (up to 19.5%) in challenging scenarios. By embedding residual-based feedback directly into the learning process, TEFL offers a simple, general, and effective enhancement to modern deep forecasting systems.

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。