arXiv 2411.173822024-11-26cs.LG

MFF-FTNet：跨频率与时间域的多尺度特征融合时间序列预测

MFF-FTNet: Multi-scale Feature Fusion across Frequency and Temporal Domains for Time Series Forecasting

Heilongjiang University（黑龙江大学）；Hangzhou Institute of Technology, Xidian University（西安电子科技大学杭州研究院）

Yangyang Shi, Qianqian Ren, Yong Liu, Jianguo Sun

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

针对时间序列预测中噪声、稀疏性及多尺度模式捕捉难题，本文提出MFF-FTNet框架，结合对比学习与跨频时多尺度特征提取，通过自适应噪声增强、双对比模块及统一特征表示，在多数据集上显著提升预测性能。

摘要：

时间序列预测在众多领域至关重要，但当前深度学习模型在噪声、数据稀疏性及捕捉复杂多尺度模式方面存在困难。本文提出MFF-FTNet，一种通过将对比学习与跨频率和时间域的多尺度特征提取相结合来应对这些挑战的新型框架。MFF-FTNet引入自适应噪声增强策略，基于原始时间序列数据的统计特性调整缩放和偏移因子，增强模型对噪声的抵抗力。该架构围绕两个互补模块构建：频率感知对比模块（FACM），通过频率选择和对比学习优化频谱表示；互补时域对比模块（CTCM），利用多尺度卷积和特征融合捕捉短期与长期依赖关系。统一特征表示策略实现跨域稳健对比学习，为精准预测提供丰富框架。在五个真实世界数据集上的大量实验表明，MFF-FTNet显著优于现有最优模型，在多变量任务上实现7.7%的MSE提升。这些发现强调了MFF-FTNet在建模复杂时间模式及处理噪声与稀疏性方面的有效性，为长短时预测提供全面解决方案。

英文摘要：

Time series forecasting is crucial in many fields, yet current deep learning models struggle with noise, data sparsity, and capturing complex multi-scale patterns. This paper presents MFF-FTNet, a novel framework addressing these challenges by combining contrastive learning with multi-scale feature extraction across both frequency and time domains. MFF-FTNet introduces an adaptive noise augmentation strategy that adjusts scaling and shifting factors based on the statistical properties of the original time series data, enhancing model resilience to noise. The architecture is built around two complementary modules: a Frequency-Aware Contrastive Module (FACM) that refines spectral representations through frequency selection and contrastive learning, and a Complementary Time Domain Contrastive Module (CTCM) that captures both short- and long-term dependencies using multi-scale convolutions and feature fusion. A unified feature representation strategy enables robust contrastive learning across domains, creating an enriched framework for accurate forecasting. Extensive experiments on five real-world datasets demonstrate that MFF-FTNet significantly outperforms state-of-the-art models, achieving a 7.7% MSE improvement on multivariate tasks. These findings underscore MFF-FTNet's effectiveness in modeling complex temporal patterns and managing noise and sparsity, providing a comprehensive solution for both long- and short-term forecasting.

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。