用新方法提升小样本多变量时间序列分类的泛化能力
COSCO: A Sharpness-Aware Training Framework for Few-shot Multivariate Time Series Classification
- 结合尖锐度感知优化与原型损失，增强模型鲁棒性
- 在少样本场景下准确率显著优于现有基线方法
- 适合数据稀缺的时间序列分类任务研究者使用
多变量时间序列分类在多个领域具有广泛应用。近年来，深度神经网络（DNN）在该任务上取得了领先性能，但通常需要大量人工标注数据，实际应用中难以满足。在小样本设置下（每类仅少量样本），DNN测试准确率显著下降，泛化能力差。本文从优化和损失函数角度出发，提出一种名为COSCO的新学习框架，包含尖锐度感知最小化（SAM）优化和原型损失函数，旨在提升DNN在小样本多变量时间序列分类中的泛化能力。实验表明，所提方法优于现有基线方法。代码已开源：https://github.com/JRB9/COSCO。
原文摘要 · Abstract (English)
Multivariate time series classification is an important task with widespread domains of applications. Recently, deep neural networks (DNN) have achieved state-of-the-art performance in time series classification. However, they often require large expert-labeled training datasets which can be infeasible in practice. In few-shot settings, i.e. only a limited number of samples per class are available in training data, DNNs show a significant drop in testing accuracy and poor generalization ability. In this paper, we propose to address these problems from an optimization and a loss function perspective. Specifically, we propose a new learning framework named COSCO consisting of a sharpness-aware minimization (SAM) optimization and a Prototypical loss function to improve the generalization ability of DNN for multivariate time series classification problems under few-shot setting. Our experiments demonstrate our proposed method outperforms the existing baseline methods. Our source code is available at: https://github.com/JRB9/COSCO.
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。