AME-TS：基于锚定的混合专家模型用于时间序列预测
AME-TS: Anchored Mixture-of-Experts for Time Series Forecasting
AI总结：
提出AME-TS，一种结构引导的稀疏时间序列基础模型，通过轻量级预测器估计序列级描述符并生成专家软结构先验，实现专家路由与可解释时间结构对齐，在GIFT-Eval基准上实现精度-效率权衡，并在M5微调中展现更稳定的专家专业化。
摘要：
时间序列预测模型通过大型Transformer骨干不断扩展规模，但大多数现有方法通过共享密集计算路径处理所有序列，尽管时间结构存在显著异质性。混合专家模型（MoE）通过条件计算提供了一种自然替代方案，但标准MoE路由导致专家专业化识别弱且在下游适应中常不稳定。我们提出AME-TS，一种结构引导的稀疏时间序列基础模型，将专家路由与可解释的时间结构对齐。AME-TS首先使用轻量级预测器估计序列级描述符，包括可预测性、季节性、趋势和稀疏性，并将其映射为专家上的软结构先验。该序列级先验在训练期间指导令牌级路由，鼓励结构对齐的专业化。在GIFT-Eval基准上，AME-TS在不同模型规模下提供了强大的精度-效率权衡：在小型模型规模上显著优于现有时间序列基础模型，在较大规模上与最强模型保持竞争力，同时通过稀疏路由激活显著更少的参数。我们进一步表明，在M5数据集微调期间，AME-TS学习了比标准MoE更可解释的路由几何和更稳定的专家专业化。这些结果表明，结构感知路由是实现稀疏专家模型在时间序列预测中优势的有效且可靠方式。
英文摘要：
Time series forecasting models are increasingly scaled through large Transformer backbones, yet most existing approaches process all series through a shared dense computation path despite substantial heterogeneity in temporal structure. Mixture-of-Experts (MoE) offers a natural alternative by enabling conditional computation, but standard MoE routing leaves expert specialization weakly identified and often unstable during downstream adaptation. We propose AME-TS, a structure-guided sparse time series foundation model that aligns expert routing with interpretable temporal structure. AME-TS first uses a lightweight regime predictor to estimate series-level descriptors, including forecastability, seasonality, trend, and sparsity, and maps them to a soft structural prior over experts. This series-level prior guides token-level routing during training, encouraging structure-aligned specialization. On the GIFT-Eval benchmark, AME-TS delivers a strong accuracy-efficiency tradeoff across model scales: it substantially outperforms existing time series foundation models at small model scales and remains competitive with the strongest models at larger scales, while activating substantially fewer parameters through sparse routing. We further show that AME-TS learns more interpretable routing geometry and substantially more stable expert specialization than standard MoE during fine-tuning on the M5 dataset. These results suggest that structure-aware routing is an effective and reliable way to realize the benefits of sparse expert models for time series forecasting.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
