用可学习的时序查询捕捉变量间全局关联，提升多变量时间序列预测精度与效率。
Temporal Query Network for Efficient Multivariate Time Series Forecasting
- 设计时序查询机制，用周期偏移的可学习向量捕捉跨变量全局模式。
- 在12个真实数据集上达到领先精度，高维数据下仍保持线性复杂度效率。
- 结构极简，仅需单层注意力+轻量MLP，适合实际部署场景。
充分建模变量间的相关性对实现高精度多变量时间序列预测（MTSF）至关重要。本文提出一种名为时序查询（Temporal Query, TQ）的新技术，更有效地捕捉多变量相关性，从而提升模型在MTSF任务中的表现。技术上，TQ利用周期性偏移的可学习向量作为注意力机制中的查询，以捕捉全局跨变量模式；而键和值则来自原始输入数据，用于编码局部样本级相关性。基于TQ技术，我们构建了一个简单且高效的模型——时序查询网络（TQNet），其仅采用单层注意力机制和轻量级多层感知机（MLP）。大量实验表明，TQNet能学习到更鲁棒的多变量相关性，在12个具有挑战性的真实世界数据集上达到当前最优预测精度。此外，即使在高维数据上，TQNet也具备与线性方法相当的高效性，平衡了性能与计算成本。代码已开源：https://github.com/ACAT-SCUT/TQNet。
原文摘要 · Abstract (English)
Sufficiently modeling the correlations among variables (aka channels) is crucial for achieving accurate multivariate time series forecasting (MTSF). In this paper, we propose a novel technique called Temporal Query (TQ) to more effectively capture multivariate correlations, thereby improving model performance in MTSF tasks. Technically, the TQ technique employs periodically shifted learnable vectors as queries in the attention mechanism to capture global inter-variable patterns, while the keys and values are derived from the raw input data to encode local, sample-level correlations. Building upon the TQ technique, we develop a simple yet efficient model named Temporal Query Network (TQNet), which employs only a single-layer attention mechanism and a lightweight multi-layer perceptron (MLP). Extensive experiments demonstrate that TQNet learns more robust multivariate correlations, achieving state-of-the-art forecasting accuracy across 12 challenging real-world datasets. Furthermore, TQNet achieves high efficiency comparable to linear-based methods even on high-dimensional datasets, balancing performance and computational cost. The code is available at: https://github.com/ACAT-SCUT/TQNet.
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。