用小波变换增强专家混合模型，提升时间序列预测精度。
WaveMoE: A Wavelet-Enhanced Mixture-of-Experts Foundation Model for Time Series Forecasting
- 双路径架构同步处理时序与小波特征，共享路由机制实现专家专精。
- 在16个基准数据集上验证，融合小波域数据显著提升预测性能。
- 适合需要高精度时序建模的场景，如金融、气象与工业预测。
时间序列基础模型（TSFMs）近年来通过在多样化时间序列数据上进行大规模预训练，在通用预测任务中取得显著成果。与此同时，引入频域信息有助于增强对复杂时序模式（如周期性、局部高频动态）的建模能力，这些模式在真实世界时间序列中广泛存在。为推动这一方向，我们提出一种新视角：将显式的频域表示融入可扩展的基础模型，并提出WaveMoE——一种用于时间序列预测的小波增强型专家混合基础模型。WaveMoE采用双路径架构，沿统一时间轴联合处理时序标记与小波标记，并通过共享专家路由机制协调二者，实现一致的专家专精并高效扩展模型容量。初步实验在16个多样化的基准数据集上表明，融合小波域语料有望进一步提升预测性能。
原文摘要 · Abstract (English)
Time series foundation models (TSFMs) have recently achieved remarkable success in universal forecasting by leveraging large-scale pretraining on diverse time series data. Complementing this progress, incorporating frequency-domain information yields promising performance in enhancing the modeling of complex temporal patterns, such as periodicity and localized high-frequency dynamics, which are prevalent in real-world time series. To advance this direction, we propose a new perspective that integrates explicit frequency-domain representations into scalable foundation models, and introduce WaveMoE, a wavelet-enhanced mixture-of-experts foundation model for time series forecasting. WaveMoE adopts a dual-path architecture that jointly processes time series tokens and wavelet tokens aligned along a unified temporal axis, and coordinates them through a shared expert routing mechanism that enables consistent expert specialization while efficiently scaling model capacity. Preliminary experimental results on 16 diverse benchmark datasets indicate that WaveMoE has the potential to further improve forecasting performance by incorporating wavelet-domain corpora.
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。