提出基于真实标签的时序预测测试时自适应新范式
Towards Principled Test-Time Adaptation for Time Series Forecasting
- 用已知真实值构建更清晰的测试时自适应协议
- 在多个数据集上表现稳定，参数量显著减少
- 适合需要轻量级、可靠自适应的时序预测场景
测试时自适应（TTA）近年成为改善时序预测（TSF）在分布偏移下性能的有前景方法。现有TSF-TTA方法对已知目标值的利用方式各异，导致适应协议不统一且缺乏明确规范。为此，我们从协议清晰性角度重新审视TSF-TTA，提出仅依赖成熟真实标签的适应协议，建立更严谨的适应设定。在此基础上，我们通过频域分析诊断现有适配器，发现其预测修正常表现为有限且结构弱的频谱变化。受此启发，我们提出频率感知校准（FAC），一种直接在频域参数化预测修正的轻量级校准方法。在多种数据集、预测跨度及源预测器下，FAC均实现竞争力与一致性表现，且可训练参数远少于对比方法。
原文摘要 · Abstract (English)
Test-time adaptation (TTA) has recently emerged as a promising approach for improving time series forecasting (TSF) under distribution shift. Existing TSF-TTA methods differ in how they utilize revealed targets, yet the resulting adaptation protocols remain heterogeneous and lack a clearly unified formulation. To address this issue, we revisit TSF-TTA from the perspective of protocol cleanliness and propose an adaptation protocol based solely on matured ground truth, yielding a more principled setting for adaptation. Under this protocol, we further diagnose existing adapters in the frequency domain and find that their prediction corrections often exhibit limited and weakly structured spectral modifications. Motivated by this diagnosis, we propose Frequency-Aware Calibration (FAC), a lightweight calibration method that directly parameterizes prediction corrections in the frequency domain. Across diverse datasets, forecasting horizons, and source forecasters, FAC achieves competitive and consistent performance while requiring substantially fewer trainable parameters than the compared TSF-TTA adapters.
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。