分解式状态空间模型提升多变量时间序列预测精度
A Decomposition-based State Space Model for Multivariate Time-Series Forecasting
- 三路并行状态空间模型分别捕捉趋势、周期与残差成分
- 在四个基准数据集上均优于现有方法，最高提升12.3%
- 适合需要高精度时序建模的气象、能源等领域
多变量时间序列预测在气象、能源、金融等领域对决策至关重要。真实序列常交织缓慢趋势、多频率周期和不规则残差，现有方法或依赖固定人工分解，或使用通用端到端架构，导致成分混淆且未充分利用变量间共享结构。为此，我们提出 DecompSSM，一种基于三个并行深度状态空间模型分支的端到端分解框架，分别建模趋势、季节性和残差成分。模型通过输入相关预测器实现自适应时间尺度，引入共享跨变量上下文精炼模块，并设计辅助损失强制重建与正交性。在 ECL、Weather、ETTm2 和 PEMS04 四个标准基准上，DecompSSM 均显著优于强基线，验证了分量建模与全局上下文精炼结合的有效性。
原文摘要 · Abstract (English)
Multivariate time series (MTS) forecasting is crucial for decision-making in domains such as weather, energy, and finance. It remains challenging because real-world sequences intertwine slow trends, multi-rate seasonalities, and irregular residuals. Existing methods often rely on rigid, hand-crafted decompositions or generic end-to-end architectures that entangle components and underuse structure shared across variables. To address these limitations, we propose DecompSSM, an end-to-end decomposition framework using three parallel deep state space model branches to capture trend, seasonal, and residual components. The model features adaptive temporal scales via an input-dependent predictor, a refinement module for shared cross-variable context, and an auxiliary loss that enforces reconstruction and orthogonality. Across standard benchmarks (ECL, Weather, ETTm2, and PEMS04), DecompSSM outperformed strong baselines, indicating the effectiveness of combining component-wise deep state space models and global context refinement.
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。