当前位置: X-MOL 学术Pers. Ubiquitous Comput. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Convergence of multiple deep neural networks for classification with fewer labeled data
Personal and Ubiquitous Computing Pub Date : 2020-08-28 , DOI: 10.1007/s00779-020-01448-6
Chuho Yi , Jungwon Cho

With the advent of deep neural networks (DNNs) in the last two decades, tremendous developments have been made in many fields, such as image classification/recognition, voice recognition, and action recognition. These advanced DNNs require large amounts of labeled data, whose collection is costly and requires great effort. In this paper, we provide a convergence method for DNNs to solve some of these difficulties. First, we consider how to create labeled data using a generative adversarial network (GAN), one DNN method, and add additional networks to improve the quality of generated data. Then, we propose a convergence method for the DNNs and use a three-step evaluation to confirm this approach and show how to use the automatically generated data for training. With the method proposed in this paper, we hope that the manual work of labeling data can be reduced for many DNN applications.



中文翻译:

多个深度神经网络的收敛性,以减少标记数据

在过去的二十年中,随着深度神经网络(DNN)的出现,图像分类/识别,语音识别和动作识别等许多领域都取得了长足的发展。这些高级DNN需要大量带标签的数据,这些数据的收集成本高昂并且需要付出巨大的努力。在本文中,我们为DNN提供了一种收敛方法来解决其中的一些困难。首先,我们考虑如何使用一种生成对抗网络(GAN)和一种DNN方法来创建标记数据,并添加其他网络以提高生成数据的质量。然后,我们提出了DNN的收敛方法,并使用三步评估来确认这种方法,并展示如何使用自动生成的数据进行训练。使用本文提出的方法,

更新日期:2020-08-28
down
wechat
bug