预训练预测模型可作时间序列分类的通用特征提取器。
Pre-trained Forecasting Models: Strong Zero-Shot Feature Extractors for Time Series Classification
- 冻结预训练预测模型提取特征，无需微调。
- 性能媲美甚至超过专为分类设计的模型。
- 预测能力越强，分类表现越好，适合通用建模。
近期时间序列基础模型研究多聚焦于预测任务，但其学习表征的泛化能力尚不明确。本文探究冻结的预训练预测模型能否有效支持分类任务。通过比较不同特征提取策略，并引入两种模型无关的嵌入增强方法，实验表明最优预测模型在分类任务上达到与当前最先进分类模型相当甚至更优的准确率。此外，预测性能与分类性能呈正相关。结果挑战了任务特定预训练的必要性，表明学习预测可能是构建通用时间序列基础模型的有效路径。
原文摘要 · Abstract (English)
Recent research on time series foundation models has primarily focused on forecasting, leaving it unclear how generalizable their learned representations are. In this study, we examine whether frozen pre-trained forecasting models can provide effective representations for classification. To this end, we compare different representation extraction strategies and introduce two model-agnostic embedding augmentations. Our experiments show that the best forecasting models achieve classification accuracy that matches or even surpasses that of state-of-the-art models pre-trained specifically for classification. Moreover, we observe a positive correlation between forecasting and classification performance. These findings challenge the assumption that task-specific pre-training is necessary, and suggest that learning to forecast may provide a powerful route toward constructing general-purpose time series foundation models.
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。