根据数据分布变化自动调节适应强度，提升时序预测的实时适应能力。
RG-TTA: Regime-Guided Meta-Control for Test-Time Adaptation in Streaming Time Series
- 用多种统计指标评估数据分布相似性，动态调整学习率和梯度更新力度。
- 在156组实验中表现最优，比普通TTA降低5.7%误差，速度提升5.5%。
- 适用于各类时序模型，适合需要持续在线适应的工业级预测场景。
测试时自适应（TTA）使神经预测模型能在流式时序数据中应对分布漂移，但现有方法对所有漂移一视同仁。本文提出制度引导的测试时自适应（RG-TTA），通过元控制器依据当前批次与历史制度的分布相似性，连续调节适应强度。利用柯尔莫哥洛夫-斯米尔诺夫、瓦舍尔斯坦-1、特征距离与方差比四类度量组成集成，计算相似性得分，并据此（i）平滑调节学习率——对新分布更激进，对熟悉分布更保守；（ii）通过损失驱动的早停控制梯度投入，而非固定预算，实现按需分配。作为补充机制，RG-TTA通过门控机制复用制度记忆中的检查点，仅当存储的专业模型性能提升≥30%时才加载。该方法具备模型无关性和策略可组合性，可封装任意支持训练/预测/保存/加载接口的预测器，并增强任意基于梯度的TTA方法。我们验证了三种组合：RG-TTA、RG-EWC、RG-DynaTTA，共比较6种更新策略（3基线+3制度引导变体），在4种紧凑架构（GRU、iTransformer、PatchTST、DLinear）、14个数据集（6个真实多变量基准+8个合成制度场景）、4个预测时长（96, 192, 336, 720）下，采用流式评估协议并运行3随机种子（总计672次实验）。制度引导策略在224组种子平均实验中取得最低均方误差（MSE）达156次（69.6%），其中RG-EWC占30.4%，RG-TTA占29.0%。总体而言，RG-TTA相比传统TTA降低MSE 5.7%，运行速度提升5.5%；RG-EWC相比独立EWC降低MSE 14.1%。
原文摘要 · Abstract (English)
Test-time adaptation (TTA) enables neural forecasters to adapt to distribution shifts in streaming time series, but existing methods apply the same adaptation intensity regardless of the nature of the shift. We propose Regime-Guided Test-Time Adaptation (RG-TTA), a meta-controller that continuously modulates adaptation intensity based on distributional similarity to previously-seen regimes. Using an ensemble of Kolmogorov-Smirnov, Wasserstein-1, feature-distance, and variance-ratio metrics, RG-TTA computes a similarity score for each incoming batch and uses it to (i) smoothly scale the learning rate -- more aggressive for novel distributions, conservative for familiar ones -- and (ii) control gradient effort via loss-driven early stopping rather than fixed budgets, allowing the system to allocate exactly the effort each batch requires. As a supplementary mechanism, RG-TTA gates checkpoint reuse from a regime memory, loading stored specialist models only when they demonstrably outperform the current model (loss improvement >= 30%). RG-TTA is model-agnostic and strategy-composable: it wraps any forecaster exposing train/predict/save/load interfaces and enhances any gradient-based TTA method. We demonstrate three compositions -- RG-TTA, RG-EWC, and RG-DynaTTA -- and evaluate 6 update policies (3 baselines + 3 regime-guided variants) across 4 compact architectures (GRU, iTransformer, PatchTST, DLinear), 14 datasets (6 real-world multivariate benchmarks + 8 synthetic regime scenarios), and 4 forecast horizons (96, 192, 336, 720) under a streaming evaluation protocol with 3 random seeds (672 experiments total). Regime-guided policies achieve the lowest MSE in 156 of 224 seed-averaged experiments (69.6%), with RG-EWC winning 30.4% and RG-TTA winning 29.0%. Overall, RG-TTA reduces MSE by 5.7% vs TTA while running 5.5% faster; RG-EWC reduces MSE by 14.1% vs standalone EWC.
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。