Study of Restrained Network Structures for Wasserstein Generative Adversarial Networks (WGANs) on Numeric Data Augmentation

doi:10.1109/ACCESS.2020.2993839

	Study of Restrained Network Structures for Wasserstein Generative Adversarial Networks (WGANs) on Numeric Data Augmentation
	Wang, Wei 1,3; Wang, Chuang 1; Cui, Tao 2; Li, Yue 1,3
刊名	IEEE ACCESS
	2020
卷号	8 页码:89812-89821
关键词	Gallium nitride Generative adversarial networks Tensile stress Generators Indexes Training Numerical models Restrained network structures generative adversarial network numeric data augmentation
ISSN号	2169-3536
DOI	10.1109/ACCESS.2020.2993839
英文摘要	Some recent studies have suggested using Generative Adversarial Network (GAN) for numeric data over-sampling, which is to generate data for completing the imbalanced numeric data. Compared with the conventional over-sampling methods, taken SMOTE as an example, the recently-proposed GAN schemes fail to generate distinguishable augmentation results for classifiers. In this paper, we discuss the reason for such failures, based on which we further study the restrained conditions between $G$ and $D$ theoretically, and propose a quantitative indicator of the restrained structure, called Similarity of the Restrained Condition (SRC) to measure the restrained conditions. Practically, we propose several candidate solutions, which are isomorphic (IWGAN) mirror (MWGAN) and self-symmetric WGAN (SWGAN) for restrained conditions. Besides, the restrained WGANs enhance the classification performance in AUC on five classifiers compared with the original data as the baseline, conventional SMOTE, and other GANs add up to 20 groups of experiments in four datasets. The restrained WGANs outperform all others in 17/20 groups, among which IWGAN accounted for 15/17 groups and the SRC is an effective measure in evaluating the restraints so that further GAN structures with G-D restrains could be designed on SRC. Multidimensional scaling (MDS) is introduced to eliminate the impact of datasets and evaluation of the AUC in a composite index and IWGAN decreases the MDS distance by 20 & x0025; to 40 & x0025;. Moreover, the convergence speed of IWGAN is increased, and the initial error of loss function is reduced.
资助项目	Tianjin Natural Science Foundations[17JCYBJC23000] ; National Key Research and Development Program of China[2018hjyzkfkt-002] ; National Key Research and Development Program of China[2016YFB0201304] ; Fundamental Research Funds for the Central Universities, Nankai University[070/63191114]
WOS研究方向	Computer Science ; Engineering ; Telecommunications
语种	英语
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
WOS记录号	WOS:000538727700078
内容类型	期刊论文
源URL	[http://ir.amss.ac.cn/handle/2S8OKBNM/51654]
专题	中国科学院数学与系统科学研究院
通讯作者	Li, Yue
作者单位	1.Nankai Univ, Coll Comp Sci, Tianjin 300350, Peoples R China 2.Chinese Acad Sci, Acad Math & Syst Sci, Natl Ctr Math & Interdisciplinary Sci, State Key Lab Sci & Engn Comp, Beijing 100190, Peoples R China 3.Key Lab Med Data Anal & Stat Res Tianjin KLMDASR, Tianjin 300350, Peoples R China
推荐引用方式 GB/T 7714	Wang, Wei,Wang, Chuang,Cui, Tao,et al. Study of Restrained Network Structures for Wasserstein Generative Adversarial Networks (WGANs) on Numeric Data Augmentation[J]. IEEE ACCESS,2020,8:89812-89821.
APA	Wang, Wei,Wang, Chuang,Cui, Tao,&Li, Yue.(2020).Study of Restrained Network Structures for Wasserstein Generative Adversarial Networks (WGANs) on Numeric Data Augmentation.IEEE ACCESS,8,89812-89821.
MLA	Wang, Wei,et al."Study of Restrained Network Structures for Wasserstein Generative Adversarial Networks (WGANs) on Numeric Data Augmentation".IEEE ACCESS 8(2020):89812-89821.