Circular Complex-Valued GMDH-Type Neural Network for Real-Valued Classification Problems

doi:10.1109/TNNLS.2020.2966031

	Circular Complex-Valued GMDH-Type Neural Network for Real-Valued Classification Problems
	Xiao, Jin 2; Jia, Yanlin 2; Jiang, Xiaoyi 1; Wang, Shouyang 3
刊名	IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
	2020-12-01
卷号	31 期号:12 页码:5285-5299
关键词	Biological system modeling Data models Biological neural networks Predictive models Neurons Mathematical model Complex-valued external criterion complex-valued group method of data handling (GMDH)-type neural network parameter estimation real-valued classification
ISSN号	2162-237X
DOI	10.1109/TNNLS.2020.2966031
英文摘要	Recently, applications of complex-valued neural networks (CVNNs) to real-valued classification problems have attracted significant attention. However, most existing CVNNs are black-box models with poor explanation performance. This study extends the real-valued group method of data handling (RGMDH)-type neural network to the complex field and constructs a circular complex-valued group method of data handling (C-CGMDH)-type neural network, which is a white-box model. First, a complex least squares method is proposed for parameter estimation. Second, a new complex-valued symmetric regularity criterion is constructed with a logarithmic function to represent explicitly the magnitude and phase of the actual and predicted complex output to evaluate and select the middle candidate models. Furthermore, the property of this new complex-valued external criterion is proven to be similar to that of the real external criterion. Before training this model, a circular transformation is used to transform the real-valued input features to the complex field. Twenty-five real-valued classification data sets from the UCI Machine Learning Repository are used to conduct the experiments. The results show that both RGMDH and C-CGMDH models can select the most important features from the complete feature space through a self-organizing modeling process. Compared with RGMDH, the C-CGMDH model converges faster and selects fewer features. Furthermore, its classification performance is statistically significantly better than the benchmark complex-valued and real-valued models. Regarding time complexity, the C-CGMDH model is comparable with other models in dealing with the data sets that have few features. Finally, we demonstrate that the GMDH-type neural network can be interpretable.
资助项目	Major Project of the National Social Science Foundation of China[18VZL006] ; EU Horizon 2020 RISE Project ULTRACEPT[778062] ; National Natural Science Foundation of China[71471124] ; Tianfu TenThousand Talents Program of Sichuan Province ; Excellent Youth Fund of Sichuan University[skqx201607] ; Excellent Youth Fund of Sichuan University[sksyl201709] ; Excellent Youth Fund of Sichuan University[skzx2016-rcrw14] ; Leading Cultivation Talents Program of Sichuan University ; Teacher and Student Joint Innovation Project of Business School of Sichuan University[LH2018011]
WOS研究方向	Computer Science ; Engineering
语种	英语
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
WOS记录号	WOS:000595533300020
内容类型	期刊论文
源URL	[http://ir.amss.ac.cn/handle/2S8OKBNM/57840]
专题	中国科学院数学与系统科学研究院
通讯作者	Wang, Shouyang
作者单位	1.Univ Munster, Fac Math & Comp Sci, D-48149 Munster, Germany 2.Sichuan Univ, Business Sch, Chengdu 610064, Peoples R China 3.Chinese Acad Sci, Acad Math & Syst Sci, Beijing 100190, Peoples R China
推荐引用方式 GB/T 7714	Xiao, Jin,Jia, Yanlin,Jiang, Xiaoyi,et al. Circular Complex-Valued GMDH-Type Neural Network for Real-Valued Classification Problems[J]. IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS,2020,31(12):5285-5299.
APA	Xiao, Jin,Jia, Yanlin,Jiang, Xiaoyi,&Wang, Shouyang.(2020).Circular Complex-Valued GMDH-Type Neural Network for Real-Valued Classification Problems.IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS,31(12),5285-5299.
MLA	Xiao, Jin,et al."Circular Complex-Valued GMDH-Type Neural Network for Real-Valued Classification Problems".IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS 31.12(2020):5285-5299.