不规则时间序列中采样模式偏移下的域泛化
Domain Generalization under Sampling Pattern Shifts in Irregular Time Series
浏览论文内容
中文总结 AI 辅助
针对不规则采样时间序列中的采样模式偏移，提出首个基准HAR-C和DG框架PRISM，通过互补表示学习与稳健训练提升对未见采样偏移的鲁棒性。
中文摘要 AI 辅助
不规则采样的多元时间序列（ISMTS）在现实应用中普遍存在，其中观测时间和可用测量在不同域之间可能有显著差异。尽管最近的模型越来越多地利用此类采样信息进行预测，但其在采样模式偏移下的鲁棒性仍未得到充分探索。我们引入了HAR-C，据我们所知，这是首个针对ISMTS中采样模式偏移的受控基准，并表明仅采样偏移就能显著降低性能，诱发采样特定的捷径，并且对现有的域泛化（DG）方法仍具挑战性。受这些发现启发，我们提出了PRISM，一个DG框架，首先在无任务标签的情况下学习互补的以特征为中心和以采样为中心的表示，随后在多样化的采样变化中进行稳健的监督训练，以抑制脆弱的捷径依赖。在受控和真实世界的ISMTS基准上的广泛实验表明，PRISM在应对未见过的采样偏移方面持续优于现有方法。我们的代码可在该URL获取。
英文摘要
Irregularly sampled multivariate time series (ISMTS) are prevalent in real-world applications, where both observation times and available measurements can vary substantially across domains. While recent models increasingly exploit such sampling information for prediction, its robustness under sampling pattern shifts remains underexplored. We introduce HAR-C, to the best of our knowledge the first controlled benchmark for sampling pattern shifts in ISMTS, and show that sampling shifts alone can substantially degrade performance, induce sampling-specific shortcuts, and remain challenging for existing domain generalization (DG) methods. Motivated by these findings, we propose PRISM, a DG framework that first learns complementary feature-centric and sampling-centric representations without task labels, and subsequently performs robust supervised training across diverse sampling variations to discourage brittle shortcut reliance. Extensive experiments on controlled and real-world ISMTS benchmarks demonstrate that PRISM consistently improves robustness to unseen sampling shifts over existing methods. Our code is available at https://anonymous.4open.science/r/PRISM.
