提出细粒度时空图滤波方法，动态保留关键依赖关系。
TimeFilter: Patch-Specific Spatial-Temporal Graph Filtration for Time Series Forecasting
- 基于图神经网络构建时空依赖，按局部片段自适应过滤无关连接。
- 在13个真实数据集上达到顶尖性能，显著优于现有方法。
- 适合需要捕捉复杂时变依赖的时序预测任务。
时序预测方法主要分为通道独立（CI）和通道依赖（CD）两类。CI忽略重要协变量关系，而CD无差别建模所有依赖，引入噪声并降低泛化能力。近期通道聚类（CC）通过分组相似通道并应用定制建模来优化依赖建模，但粗粒度聚类难以有效捕捉复杂且随时间变化的交互。为此，我们提出TimeFilter，一种基于图神经网络的自适应、细粒度依赖建模框架。在从输入序列构建图后，TimeFilter通过在片段特定（patch-specific）基础上过滤无关相关性，同时保留最关键的关系，以精炼学习到的时空依赖。在涵盖多个应用领域的13个真实世界数据集上的大量实验表明，TimeFilter表现达到当前最优水平。代码已开源：https://github.com/TROUBADOUR000/TimeFilter。
原文摘要 · Abstract (English)
Time series forecasting methods generally fall into two main categories: Channel Independent (CI) and Channel Dependent (CD) strategies. While CI overlooks important covariate relationships, CD captures all dependencies without distinction, introducing noise and reducing generalization. Recent advances in Channel Clustering (CC) aim to refine dependency modeling by grouping channels with similar characteristics and applying tailored modeling techniques. However, coarse-grained clustering struggles to capture complex, time-varying interactions effectively. To address these challenges, we propose TimeFilter, a GNN-based framework for adaptive and fine-grained dependency modeling. After constructing the graph from the input sequence, TimeFilter refines the learned spatial-temporal dependencies by filtering out irrelevant correlations while preserving the most critical ones in a patch-specific manner. Extensive experiments on 13 real-world datasets from diverse application domains demonstrate the state-of-the-art performance of TimeFilter. The code is available at https://github.com/TROUBADOUR000/TimeFilter.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。