期刊
NEURAL NETWORKS
卷 165, 期 -, 页码 868-883出版社
PERGAMON-ELSEVIER SCIENCE LTD
DOI: 10.1016/j.neunet.2023.06.033
关键词
Time series; Classification; Clustering; Generative Adversarial Networks; Deep Neural Networks; Representation learning
Recent research has shown that supervised Convolutional Neural Networks (CNNs) are superior in learning hierarchical representations from time series data for successful classification. However, obtaining high-quality labeled time series data can be costly and infeasible. This study introduces a Time-series Convolutional GAN (TCGAN), which uses a generator and a discriminator to play an adversarial game and learns representations for time series recognition without label information. Experimental results demonstrate that TCGAN is faster and more accurate than existing time-series GANs, and the learned representations enable simple classification and clustering methods to achieve superior performance in scenarios with few-labeled and imbalanced-labeled data.
Recent works have demonstrated the superiority of supervised Convolutional Neural Networks (CNNs) in learning hierarchical representations from time series data for successful classification. These meth-ods require sufficiently large labeled data for stable learning, however acquiring high-quality labeled time series data can be costly and potentially infeasible. Generative Adversarial Networks (GANs) have achieved great success in enhancing unsupervised and semi-supervised learning. Nonetheless, to our best knowledge, it remains unclear how effectively GANs can serve as a general-purpose solution to learn representations for time series recognition, i.e., classification and clustering. The above considerations inspire us to introduce a Time-series Convolutional GAN (TCGAN). TCGAN learns by playing an adversarial game between two one-dimensional CNNs (i.e., a generator and a discriminator) in the absence of label information. Parts of the trained TCGAN are then reused to construct a representation encoder to empower linear recognition methods. We conducted comprehensive experiments on synthetic and real-world datasets. The results demonstrate that TCGAN is faster and more accurate than existing time-series GANs. The learned representations enable simple classification and clustering methods to achieve superior and stable performance. Furthermore, TCGAN retains high efficacy in scenarios with few-labeled and imbalanced-labeled data. Our work provides a promising path to effectively utilize abundant unlabeled time series data. & COPY; 2023 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据