CASIA OpenIR
Kernelized support vector machine with deep learning: An efficient approach for extreme multiclass dataset
Zareapoor, Masoumeh1; Shamsolmoali, Pourya1; Jain, Deepak Kumar2; Wanx, Haoxiang3,4; Yang, Jie1
Source PublicationPATTERN RECOGNITION LETTERS
ISSN0167-8655
2018-11-01
Volume115Pages:4-13
Corresponding AuthorYang, Jie(jieyang@sjtu.edu.cn)
AbstractClassification with thousands of classes and a large number of features is often computationally intractable. The presence of irrelevant features can decrease the classification performance and increase the computational complexity of classification. Moreover, classification with many classes (thousands or more) often leads to class-confusability, and the more confusable classes increase the training error. A Robust classification model for use in high-dimensional data with a large number of classes (e.g. k >= 10(4) ) requires a prudent combination of a feature extractor and a classifier. While support vector machines with the appropriate kernel is promising regarding producing decision from the well-behaved features, but often present a negative repercussion at modeling the large scale data and more especially, the ultra-large number of classes. Architectures such as deep belief networks exhibit an impressive power to learn and collect robust features. In this paper, we present a hybrid system where a supervised deep belief network is trained to select generic features, and a kernel-based SVM is trained from the features that learned by the DBN. In our hybrid model, we substituted linear kernel for nonlinear ones (due to a large number of classes) without loss of accuracy, and gives significant gains on real world dataset involving 20,000 to 65,000 classes compared to state of the art approaches. (C) 2017 Elsevier B.V. All rights reserved.
KeywordDeep learning Deep belief networks Multiclass-classification Kernel-based SVM Feature extraction
DOI10.1016/j.patrec.2017.09.018
WOS KeywordCLASSIFICATION ; SVM
Indexed BySCI
Language英语
WOS Research AreaComputer Science
WOS SubjectComputer Science, Artificial Intelligence
WOS IDWOS:000451025900002
PublisherELSEVIER SCIENCE BV
Citation statistics
Cited Times:9[WOS]   [WOS Record]     [Related Records in WOS]
Document Type期刊论文
Identifierhttp://ir.ia.ac.cn/handle/173211/25688
Collection中国科学院自动化研究所
Corresponding AuthorYang, Jie
Affiliation1.Shanghai Jiao Tong Univ, Inst Image Proc & Pattern Recognit, Shanghai 200240, Peoples R China
2.Chinese Acad Sci, Inst Automat, Beijing, Peoples R China
3.Cornell Univ, Dept ECE, Ithaca, NY 14853 USA
4.GoPercept Lab, R&D Ctr, Ithaca, NY USA
Recommended Citation
GB/T 7714
Zareapoor, Masoumeh,Shamsolmoali, Pourya,Jain, Deepak Kumar,et al. Kernelized support vector machine with deep learning: An efficient approach for extreme multiclass dataset[J]. PATTERN RECOGNITION LETTERS,2018,115:4-13.
APA Zareapoor, Masoumeh,Shamsolmoali, Pourya,Jain, Deepak Kumar,Wanx, Haoxiang,&Yang, Jie.(2018).Kernelized support vector machine with deep learning: An efficient approach for extreme multiclass dataset.PATTERN RECOGNITION LETTERS,115,4-13.
MLA Zareapoor, Masoumeh,et al."Kernelized support vector machine with deep learning: An efficient approach for extreme multiclass dataset".PATTERN RECOGNITION LETTERS 115(2018):4-13.
Files in This Item:
There are no files associated with this item.
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Zareapoor, Masoumeh]'s Articles
[Shamsolmoali, Pourya]'s Articles
[Jain, Deepak Kumar]'s Articles
Baidu academic
Similar articles in Baidu academic
[Zareapoor, Masoumeh]'s Articles
[Shamsolmoali, Pourya]'s Articles
[Jain, Deepak Kumar]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Zareapoor, Masoumeh]'s Articles
[Shamsolmoali, Pourya]'s Articles
[Jain, Deepak Kumar]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.