用频谱可预测性快速判断时间序列模型是否值得用大模型。
Spectral Predictability as a Fast Reliability Indicator for Time Series Forecasting Model Selection
- 引入频谱可预测性Ω，仅需几秒即可评估数据难度。
- Ω高时大模型显著优于小模型，Ω低时优势消失。
- 适合想快速筛选模型、避免盲目试错的从业者。
实际应用中，部署时间序列预测模型面临两难：验证数十个模型计算成本过高，但选错模型又会导致性能不佳。我们发现，频谱可预测性Ω——一种简单的信号处理指标——能系统性地划分模型族的性能表现，实现快速模型选择。我们在四个不同领域开展受控实验，并进一步扩展到51个模型和28个来自GIFT-Eval基准的数据集。结果表明，当Ω较高时，大型时间序列基础模型（TSFMs）系统性优于轻量级任务训练基线，但随着Ω下降，其优势逐渐消失。计算Ω每数据集仅需数秒，使从业者能快速判断数据是否适合使用TSFM，或只需简单便宜的模型即可。我们证明Ω能可预测地分层模型性能，提供实用的初筛工具，降低验证成本，同时凸显在真正困难（低Ω）问题上提升模型能力的必要性。
原文摘要 · Abstract (English)
Practitioners deploying time series forecasting models face a dilemma: exhaustively validating dozens of models is computationally prohibitive, yet choosing the wrong model risks poor performance. We show that spectral predictability~$Ω$ -- a simple signal processing metric -- systematically stratifies model family performance, enabling fast model selection. We conduct controlled experiments in four different domains, then further expand our analysis to 51 models and 28 datasets from the GIFT-Eval benchmark. We find that large time series foundation models (TSFMs) systematically outperform lightweight task-trained baselines when $Ω$ is high, while their advantage vanishes as $Ω$ drops. Computing $Ω$ takes seconds per dataset, enabling practitioners to quickly assess whether their data suits TSFM approaches or whether simpler, cheaper models suffice. We demonstrate that $Ω$ stratifies model performance predictably, offering a practical first-pass filter that reduces validation costs while highlighting the need for models that excel on genuinely difficult (low-$Ω$) problems rather than merely optimizing easy ones.
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。