SwitchPFN：用于冻结上下文时间序列分类的共享切换动力学

中文摘要 AI 辅助

表格基础模型（TFMs）为时间序列分类提供了一条有前景的途径，但其有效性取决于序列数据如何转换为表格表示。现有表示面临两个挑战：全局聚合可能丢失时间演化的顺序，而在独立拟合的坐标系中计算的特征可能在不同序列间缺乏一致的含义。因此，我们将TFMs的表示设计视为一个独立的问题：表示应保留局部时间转换，同时保持跨样本的共享特征定义。我们提出SwitchPFN，它从训练序列中学习共享投影和状态码本，使局部动态算子和转换特征在不同样本间直接可比。在评估的基准上，SwitchPFN在所评估方法中取得了最高的平均准确率，相对于最强基线相对提升了4.47%。消融研究、参数敏感性分析和减少训练数据的实验进一步检验了表示的贡献、其主要设计选择以及在标记数据有限时的行为。

英文摘要

Tabular foundation models (TFMs) provide a promising route to time-series classification, but their effectiveness depends on how sequential data are converted into tabular representations. Existing representations face two challenges: global aggregation can lose the order of temporal evolution, while features computed in independently fitted coordinate systems may not have consistent meanings across sequences. We therefore view representation design for TFMs as a problem in its own right: the representation should preserve local temporal transitions while maintaining a shared feature definition across samples. We propose SwitchPFN, which learns a shared projection and regime codebook from the training sequences, making local dynamic operators and transition features directly comparable across samples. Across the evaluated benchmarks, SwitchPFN achieves the highest mean accuracy among the evaluated methods, improving over the strongest baseline by 4.47% relatively. Ablation studies, parameter sensitivity analyses, and reduced-training-data experiments further examine the contributions of the representation, its main design choices, and its behavior when labeled data are limited.

发表机构

• The Chinese University of Hong Kong（香港中文大学）
• Cornell University（康奈尔大学）
• Southeast University（东南大学）
• University College London（伦敦大学学院）
• Dalian Polytechnic University（大连工业大学）
• Hainan Normal University（海南师范大学）
• Anhui Agricultural University（安徽农业大学）
• Beijing University of Posts and Telecommunications（北京邮电大学）
• East China Normal University（华东师范大学）
• Karlsruhe Institute of Technology（卡尔斯鲁厄理工学院）

机构由 AI 辅助整理，请以论文原文为准。