提出动态多尺度协同框架，提升时间序列预测精度与效率。
DMSC: Dynamic Multi-Scale Coordination Framework for Time Series Forecasting
- 动态分块自适应划分序列，实现多粒度时序特征提取
- 多层级交互建模跨变量、片内与片间依赖关系
- 自适应路由专家网络融合多尺度预测结果
时间序列预测（TSF）在建模跨尺度复杂时序依赖方面仍面临挑战。尽管已有基于CNN、MLP或Transformer的分解方法与新架构，但现有方法仍受限于静态分解策略、依赖建模碎片化及融合机制僵化，难以充分捕捉复杂时序依赖。为此，本文提出动态多尺度协同框架（DMSC），包含多尺度补丁分解模块（EMPD）、三元交互模块（TIB）和自适应尺度路由混合专家模块（ASR-MoE）。EMPD作为内置组件，动态将序列划分为指数级粒度递增的层次化补丁，通过输入自适应调整消除预设尺度约束；TIB在每层分解表示中联合建模片内、片间与跨变量依赖；EMPD与TIB构成多层渐进式级联架构，前层粗粒度表示通过门控路径自适应引导后层细粒度特征提取；ASR-MoE则利用具有时序感知加权的全局与局部专家动态融合多尺度预测结果。在13个真实世界基准上的实验表明，DMSC持续保持最先进性能并具备优异计算效率。代码已开源：https://github.com/1327679995/DMSC。
原文摘要 · Abstract (English)
Time Series Forecasting (TSF) faces persistent challenges in modeling intricate temporal dependencies across different scales. Despite recent advances leveraging different decomposition operations and novel architectures based on CNN, MLP or Transformer, existing methods still struggle with static decomposition strategies, fragmented dependency modeling, and inflexible fusion mechanisms, limiting their ability to model intricate temporal dependencies. To explicitly solve the mentioned three problems respectively, we propose a novel Dynamic Multi-Scale Coordination Framework (DMSC) with Multi-Scale Patch Decomposition block (EMPD), Triad Interaction Block (TIB) and Adaptive Scale Routing MoE block (ASR-MoE). Specifically, EMPD is designed as a built-in component to dynamically segment sequences into hierarchical patches with exponentially scaled granularities, eliminating predefined scale constraints through input-adaptive patch adjustment. TIB then jointly models intra-patch, inter-patch, and cross-variable dependencies within each layer's decomposed representations. EMPD and TIB are jointly integrated into layers forming a multi-layer progressive cascade architecture, where coarse-grained representations from earlier layers adaptively guide fine-grained feature extraction in subsequent layers via gated pathways. And ASR-MoE dynamically fuses multi-scale predictions by leveraging specialized global and local experts with temporal-aware weighting. Comprehensive experiments on thirteen real-world benchmarks demonstrate that DMSC consistently maintains state-of-the-art (SOTA) performance and superior computational efficiency for TSF tasks. Code is available at https://github.com/1327679995/DMSC.
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。