FreRA：一种用于时间序列分类对比学习的频率细化增强方法
FreRA: A Frequency-Refined Augmentation for Contrastive Learning on Time Series Classification
AI总结：
针对时间序列分类对比学习中时域增强跨模态不兼容的问题，提出频域增强方法FreRA，通过自动分离频率成分并分别修改以保护语义信息，在分类、异常检测和迁移学习任务上优于十个基线方法。
摘要：
对比学习已成为无监督表示学习的有效方法。然而，尽管最优增强策略的设计对对比学习至关重要，但在时间序列分类任务中却较少被探索。现有的预定义时域增强方法主要从视觉领域借鉴而来，并非针对时间序列数据专门设计。因此，这种跨模态不兼容性可能会通过向数据中引入不匹配的模式，从而扭曲时间序列的语义相关信息。为了解决这一局限性，我们从频域角度提出了一种新视角，并确定了其在下游分类中的三个优势：全局性、独立性和紧凑性。为了充分利用这三个特性，我们提出了一种轻量且有效的 Frequency Refined Augmentation (FreRA)，专为分类任务上的时间序列对比学习量身定制，它可以即插即用地无缝集成到对比学习框架中。具体而言，FreRA 自动分离关键和不重要的频率成分。相应地，我们提出了语义感知的 Identity Modification 和语义无关的 Self-adaptive Modification，分别用于保护关键频率成分中的语义相关信息，并向不重要的频率成分中注入方差。在理论上，我们证明了 FreRA 能生成保留语义的视图。在实验中，我们在两个基准数据集（包括 UCR 和 UEA 档案库）以及五个涵盖不同应用的大规模数据集上进行了广泛实验。FreRA 在时间序列分类、异常检测和迁移学习任务上始终优于十个领先的基线方法，证明了其在对比表示学习以及跨不同数据集的迁移学习场景中具有卓越的泛化能力。
英文摘要：
Contrastive learning has emerged as a competent approach for unsupervised representation learning. However, the design of an optimal augmentation strategy, although crucial for contrastive learning, is less explored for time series classification tasks. Existing predefined time-domain augmentation methods are primarily adopted from vision and are not specific to time series data. Consequently, this cross-modality incompatibility may distort the semantically relevant information of time series by introducing mismatched patterns into the data. To address this limitation, we present a novel perspective from the frequency domain and identify three advantages for downstream classification: global, independent, and compact. To fully utilize the three properties, we propose the lightweight yet effective Frequency Refined Augmentation (FreRA) tailored for time series contrastive learning on classification tasks, which can be seamlessly integrated with contrastive learning frameworks in a plug-and-play manner. Specifically, FreRA automatically separates critical and unimportant frequency components. Accordingly, we propose semantic-aware Identity Modification and semantic-agnostic Self-adaptive Modification to protect semantically relevant information in the critical frequency components and infuse variance into the unimportant ones respectively. Theoretically, we prove that FreRA generates semantic-preserving views. Empirically, we conduct extensive experiments on two benchmark datasets, including UCR and UEA archives, as well as five large-scale datasets on diverse applications. FreRA consistently outperforms ten leading baselines on time series classification, anomaly detection, and transfer learning tasks, demonstrating superior capabilities in contrastive representation learning and generalization in transfer learning scenarios across diverse datasets.
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。
