A Fuzzy Deep Model Based on Fuzzy Restricted Boltzmann Machines for High-Dimensional Data Classification

doi:10.1109/TFUZZ.2019.2902111

	A Fuzzy Deep Model Based on Fuzzy Restricted Boltzmann Machines for High-Dimensional Data Classification
	Feng, Shuang 4,5; Chen, C. L. Philip 2,3,4; Zhang, Chun-Yang 1
刊名	IEEE TRANSACTIONS ON FUZZY SYSTEMS
	2020-07-01
卷号	28 期号:7 页码:1344-1355
关键词	Data models Training Training data Fuzzy neural networks Databases Neural networks Classification fuzzy deep model fuzzy restricted Boltzmann machine (FRBM) hybrid learning
ISSN号	1063-6706
DOI	10.1109/TFUZZ.2019.2902111
通讯作者	Feng, Shuang(fengshuang@bnuz.edu.cn)
英文摘要	We establish a fuzzy deep model called the fuzzy deep belief net (FDBN) based on fuzzy restricted Boltzmann machines (FRBMs) due to their excellent generative and discriminative properties. The learning procedure of an FDBN is divided into a pretraining phase and a subsequent fine-tuning phase. In the pretraining phase, a group of FRBMs is trained in a greedy layerwise way: the first FRBM is trained by original samples, and the average values of the left and right probabilities produced by its hidden units are treated as the training data for subsequent FRBMs. The resulting FDBN is either a generative or a discriminative model depending on the choice of training a generative or a discriminative type of FRBM on top. Then, a hybrid learning approach is proposed to fine-tune this novel fuzzy deep model: the well pretrained fuzzy parameters are first defuzzified, and the FDBN with defuzzified parameters is fine-tuned by the wake-sleep or stochastic gradient descent algorithm. This hybrid strategy not only avoids learning an intractable fuzzy neural network, but also greatly improves the classification capability of the FDBN. The experimental results on MNIST, NORB, and 15 Scene databases indicate that the FDBN with the hybrid learning approach can handle high-dimensional raw images directly. It inherits the fine nature of the FRBM and outperforms some state-of-the-art discriminative models in classification accuracy. Moreover, it shows better capability of robustness than a deep belief net when encountering noisy data.
资助项目	National Natural Science Foundation of China[61572540] ; National Natural Science Foundation of China[61751202] ; National Natural Science Foundation of China[U1813203] ; National Natural Science Foundation of China[U1801262] ; National Natural Science Foundation of China[61751205] ; National Natural Science Foundation of China[61603096] ; Macau Science and Technology Development Fund[019/2015/A1] ; Macau Science and Technology Development Fund[079/2017/A2] ; Macau Science and Technology Development Fund[024/2015/AMJ] ; University of Macau ; Teacher Research Capacity Promotion Program of Beijing Normal University, Zhuhai
WOS关键词	POSSIBILISTIC MEAN-VALUE ; LEARNING ALGORITHMS ; RECOGNITION ; NETWORKS
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:000545205300014
资助机构	National Natural Science Foundation of China ; Macau Science and Technology Development Fund ; University of Macau ; Teacher Research Capacity Promotion Program of Beijing Normal University, Zhuhai
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/40035]
专题	离退休人员
通讯作者	Feng, Shuang
作者单位	1.Fuzhou Univ, Sch Math & Comp Sci, Fuzhou 350116, Peoples R China 2.Chinese Acad Sci, Inst Automat, State Key Lab Management & Control Complex Syst, Beijing 100080, Peoples R China 3.Dalian Maritime Univ, Dept Nav, Dalian 116026, Peoples R China 4.Univ Macau, Fac Sci & Technol, Macau 999078, Peoples R China 5.Beijing Normal Univ, Sch Appl Math, Zhuhai 519087, Peoples R China
推荐引用方式 GB/T 7714	Feng, Shuang,Chen, C. L. Philip,Zhang, Chun-Yang. A Fuzzy Deep Model Based on Fuzzy Restricted Boltzmann Machines for High-Dimensional Data Classification[J]. IEEE TRANSACTIONS ON FUZZY SYSTEMS,2020,28(7):1344-1355.
APA	Feng, Shuang,Chen, C. L. Philip,&Zhang, Chun-Yang.(2020).A Fuzzy Deep Model Based on Fuzzy Restricted Boltzmann Machines for High-Dimensional Data Classification.IEEE TRANSACTIONS ON FUZZY SYSTEMS,28(7),1344-1355.
MLA	Feng, Shuang,et al."A Fuzzy Deep Model Based on Fuzzy Restricted Boltzmann Machines for High-Dimensional Data Classification".IEEE TRANSACTIONS ON FUZZY SYSTEMS 28.7(2020):1344-1355.