提出线性复杂度Transformer，高效预测多变量时间序列
Ister: Linear Transformer for Efficient Multivariate Time Series Forecasting
- 用点积操作替代自注意力，实现线性计算复杂度
- 在多个真实数据集上达到当前最优预测精度
- 适合需要高效处理高维时间序列的场景
基于Transformer的模型在捕捉长程依赖方面显著提升了多变量时间序列预测（MTSF）性能，但其自注意力机制的二次计算复杂度限制了在高维序列上的可扩展性。为此，我们提出Inverted Seasonal-Trend Decomposition Transformer（Ister），一种兼顾预测精度与计算效率的新架构。核心是Dot-attention，一种线性复杂度注意力机制，通过逐元素点积操作建模序列间依赖，取代传统多头自注意力。此外，引入反向季节-趋势分解策略，分离周期成分，使模型聚焦于周期模式学习，提升通道对齐效果。在多个真实世界基准测试中，Ister持续取得最先进性能。代码已开源：https://github.com/macovaseas/Ister。
原文摘要 · Abstract (English)
Transformer-based models have achieved remarkable success in multivariate time series forecasting (MTSF) by capturing long-range dependencies. However, their widespread adoption is hindered by the quadratic computational complexity of self-attention, which limits scalability on high-dimensional sequences. To address this challenge, we propose the Inverted Seasonal-Trend Decomposition Transformer (Ister), a novel architecture that enhances both predictive accuracy and computational efficiency. Central to Ister is Dot-attention, a linear-complexity attention mechanism that replaces conventional multi-head self-attention with element-wise dot-product operations to model inter-series dependencies. Furthermore, we introduce an inverted seasonal-trend decomposition strategy that isolates periodic components, enabling the model to focus learning on periodic patterns, thereby improving the performance of channel alignment. Extensive experiments across several real-world benchmarks demonstrate that Ister consistently achieves state-of-the-art performance. Code is available at https://github.com/macovaseas/Ister.
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。