U-Cast：学习用于高维时间序列预测的层级结构
U-Cast: Learning Hierarchical Structures for High-Dimensional Time Series Forecasting
AI总结：
针对高维时间序列预测中的复杂层级通道关联问题，提出U-Cast架构与Time-HD基准，通过查询注意力和全秩正则化提升预测准确率与效率。
摘要：
时间序列预测（TSF）是时间序列分析中的一个核心问题。然而，随着时间序列数据集中通道的数量扩展到数千甚至更多——我们将这一场景定义为高维时间序列预测（HDTSF）——它带来了显著的新建模挑战，而这些挑战通常并非传统 TSF 研究的主要关注点。HDTSF 之所以具有挑战性，是因为通道相关性往往形成复杂且分层的模式。现有 TSF 模型要么忽略这些交互作用，要么无法随维度增长而扩展。为解决这一问题，我们提出 U-Cast，这是一种依赖通道的预测架构，通过创新的基于查询的注意力机制学习潜在的层级通道结构。为解耦高度相关的通道表示，U-Cast 在训练期间加入全秩正则化。我们还发布了 Time-HD，这是首个大规模、多样化高维数据集基准。我们的理论表明，利用跨通道信息可降低预测风险，而在 Time-HD 上的实验证明，U-Cast 在准确率和效率上均超越强基线。U-Cast 与 Time-HD 共同为未来 HDTSF 研究提供了坚实基础。
英文摘要：
Time series forecasting (TSF) is a central problem in time series analysis. However, as the number of channels in time series datasets scales to the thousands or more, a scenario we define as High-Dimensional Time Series Forecasting (HDTSF), it introduces significant new modeling challenges that are often not the primary focus of traditional TSF research. HDTSF is challenging because the channel correlation often forms complex and hierarchical patterns. Existing TSF models either ignore these interactions or fail to scale as dimensionality grows. To address this issue, we propose U-Cast, a channel-dependent forecasting architecture that learns latent hierarchical channel structures with an innovative query-based attention. To disentangle highly correlated channel representation, U-Cast adds a full-rank regularization during training. We also release Time-HD, the first benchmark of large, diverse, high-dimensional datasets. Our theory shows that exploiting cross-channel information lowers forecasting risk, and experiments on Time-HD demonstrate that U-Cast surpasses strong baselines in both accuracy and efficiency. Together, U-Cast and Time-HD provide a solid basis for future HDTSF research.
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。
