[论文解读] Cost-Sensitive Convolution based Neural Networks for Imbalanced Time-Series Classification

Yue Geng, Xinyu Luo|arXiv (Cornell University)|Jan 13, 2018

Time Series Analysis and Forecasting参考文献 33被引用 23

一句话总结

本文提出了一种代价敏感的卷积神经网络框架，以解决时间序列分类中的类别不平衡问题。通过自适应地为少数类分配更高的误分类惩罚，该方法在真实世界数据集上显著提升了模型在不平衡时间序列数据上的性能，其中代价敏感的残差网络和CNN网络在所有指标上均优于其他模型。

ABSTRACT

Some deep convolutional neural networks were proposed for time-series classification and class imbalanced problems. However, those models performed degraded and even failed to recognize the minority class of an imbalanced temporal sequences dataset. Minority samples would bring troubles for temporal deep learning classifiers due to the equal treatments of majority and minority class. Until recently, there were few works applying deep learning on imbalanced time-series classification (ITSC) tasks. Here, this paper aimed at tackling ITSC problems with deep learning. An adaptive cost-sensitive learning strategy was proposed to modify temporal deep learning models. Through the proposed strategy, classifiers could automatically assign misclassification penalties to each class. In the experimental section, the proposed method was utilized to modify five neural networks. They were evaluated on a large volume, real-life and imbalanced time-series dataset with six metrics. Each single network was also tested alone and combined with several mainstream data samplers. Experimental results illustrated that the proposed cost-sensitive modified networks worked well on ITSC tasks. Compared to other methods, the cost-sensitive convolution neural network and residual network won out in the terms of all metrics. Consequently, the proposed cost-sensitive learning strategy can be used to modify deep learning classifiers from cost-insensitive to cost-sensitive. Those cost-sensitive convolutional networks can be effectively applied to address ITSC issues.

研究动机与目标

• 为解决由于类别不平衡导致时间序列分类中少数类表现不佳的挑战。
• 开发一种代价敏感的学习策略，根据类别分布自适应地调整误分类惩罚。
• 修改现有的深度学习架构（如CNN、ResNet），使其具备代价敏感性，以提升在不平衡时间数据上的泛化能力。
• 使用多种指标在大规模、真实世界、不平衡的时间序列数据集上评估所提方法的有效性。

提出的方法

• 提出一种自适应的代价敏感学习策略，在训练过程中动态地为少数类分配更高的误分类惩罚。
• 通过在反向传播过程中集成类别特定的损失加权，修改标准深度学习模型。
• 该方法应用于五种不同的神经网络架构，包括标准CNN和残差网络。
• 基于类别频率计算类别权重，为低频类别分配更高的惩罚，以纠正梯度更新中的偏差。
• 使用改进的交叉熵损失函数，结合代价敏感加权，端到端训练该框架。
• 将该方法作为独立模型，以及与主流数据采样技术（如SMOTE、ADASYN）结合进行评估。

实验结果

研究问题

• RQ1 代价敏感学习策略能否提升深度神经网络在不平衡时间序列数据集上的分类性能？
• RQ2 与传统数据采样技术相比，所提方法在处理时间序列分类中的类别不平衡问题时表现如何？
• RQ3 在F1-score和AUC指标上，哪些深度学习架构最受益于代价敏感的改进？
• RQ4 当应用于真实世界的大规模时间序列数据时，代价敏感方法是否保持鲁棒性？
• RQ5 所提方法能否在性能上超越基线模型以及结合数据采样与深度学习的混合方法？

主要发现

• 代价敏感的卷积神经网络和残差网络在不平衡时间序列数据集的六个评估指标上均达到最高性能。
• 与未采用代价敏感学习训练的标准深度学习模型相比，所提方法在识别少数类样本方面表现更优。
• 代价敏感方法持续提升了少数类的F1-score和AUC，表明其在罕见事件上的泛化能力更强。
• 与数据采样技术结合使用时，代价敏感模型进一步提升了性能，表明二者具有互补优势。
• 自适应的代价敏感策略有效缓解了对多数类的偏差，使模型能更好地学习少数类的时间模式。