DPWSS: differentially private working set selection for training support vector machines

Support vector machine (SVM) is a robust machine learning method and is widely used in classification. However, the traditional SVM training methods may reveal personal privacy when the training data contains sensitive information. In the training process of SVMs, working set selection is a vital st...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Zhenlong Sun, Jing Yang, Xiaoye Li, Jianpei Zhang
Formato:	article
Lenguaje:	EN
Publicado:	PeerJ Inc. 2021
Materias:	Differential privacy Exponential mechanism Sequential minimal optimization Support vector machines Working set selection Electronic computers. Computer science QA75.5-76.95
Acceso en línea:	https://doaj.org/article/fdeb66e7d0c941e78d57f937dcb031d7
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:fdeb66e7d0c941e78d57f937dcb031d7
record_format	dspace
spelling	oai:doaj.org-article:fdeb66e7d0c941e78d57f937dcb031d72021-12-03T15:05:11ZDPWSS: differentially private working set selection for training support vector machines10.7717/peerj-cs.7992376-5992https://doaj.org/article/fdeb66e7d0c941e78d57f937dcb031d72021-12-01T00:00:00Zhttps://peerj.com/articles/cs-799.pdfhttps://peerj.com/articles/cs-799/https://doaj.org/toc/2376-5992Support vector machine (SVM) is a robust machine learning method and is widely used in classification. However, the traditional SVM training methods may reveal personal privacy when the training data contains sensitive information. In the training process of SVMs, working set selection is a vital step for the sequential minimal optimization-type decomposition methods. To avoid complex sensitivity analysis and the influence of high-dimensional data on the noise of the existing SVM classifiers with privacy protection, we propose a new differentially private working set selection algorithm (DPWSS) in this paper, which utilizes the exponential mechanism to privately select working sets. We theoretically prove that the proposed algorithm satisfies differential privacy. The extended experiments show that the DPWSS algorithm achieves classification capability almost the same as the original non-privacy SVM under different parameters. The errors of optimized objective value between the two algorithms are nearly less than two, meanwhile, the DPWSS algorithm has a higher execution efficiency than the original non-privacy SVM by comparing iterations on different datasets. To the best of our knowledge, DPWSS is the first private working set selection algorithm based on differential privacy.Zhenlong SunJing YangXiaoye LiJianpei ZhangPeerJ Inc.articleDifferential privacyExponential mechanismSequential minimal optimizationSupport vector machinesWorking set selectionElectronic computers. Computer scienceQA75.5-76.95ENPeerJ Computer Science, Vol 7, p e799 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Differential privacy Exponential mechanism Sequential minimal optimization Support vector machines Working set selection Electronic computers. Computer science QA75.5-76.95
spellingShingle	Differential privacy Exponential mechanism Sequential minimal optimization Support vector machines Working set selection Electronic computers. Computer science QA75.5-76.95 Zhenlong Sun Jing Yang Xiaoye Li Jianpei Zhang DPWSS: differentially private working set selection for training support vector machines
description	Support vector machine (SVM) is a robust machine learning method and is widely used in classification. However, the traditional SVM training methods may reveal personal privacy when the training data contains sensitive information. In the training process of SVMs, working set selection is a vital step for the sequential minimal optimization-type decomposition methods. To avoid complex sensitivity analysis and the influence of high-dimensional data on the noise of the existing SVM classifiers with privacy protection, we propose a new differentially private working set selection algorithm (DPWSS) in this paper, which utilizes the exponential mechanism to privately select working sets. We theoretically prove that the proposed algorithm satisfies differential privacy. The extended experiments show that the DPWSS algorithm achieves classification capability almost the same as the original non-privacy SVM under different parameters. The errors of optimized objective value between the two algorithms are nearly less than two, meanwhile, the DPWSS algorithm has a higher execution efficiency than the original non-privacy SVM by comparing iterations on different datasets. To the best of our knowledge, DPWSS is the first private working set selection algorithm based on differential privacy.
format	article
author	Zhenlong Sun Jing Yang Xiaoye Li Jianpei Zhang
author_facet	Zhenlong Sun Jing Yang Xiaoye Li Jianpei Zhang
author_sort	Zhenlong Sun
title	DPWSS: differentially private working set selection for training support vector machines
title_short	DPWSS: differentially private working set selection for training support vector machines
title_full	DPWSS: differentially private working set selection for training support vector machines
title_fullStr	DPWSS: differentially private working set selection for training support vector machines
title_full_unstemmed	DPWSS: differentially private working set selection for training support vector machines
title_sort	dpwss: differentially private working set selection for training support vector machines
publisher	PeerJ Inc.
publishDate	2021
url	https://doaj.org/article/fdeb66e7d0c941e78d57f937dcb031d7
work_keys_str_mv	AT zhenlongsun dpwssdifferentiallyprivateworkingsetselectionfortrainingsupportvectormachines AT jingyang dpwssdifferentiallyprivateworkingsetselectionfortrainingsupportvectormachines AT xiaoyeli dpwssdifferentiallyprivateworkingsetselectionfortrainingsupportvectormachines AT jianpeizhang dpwssdifferentiallyprivateworkingsetselectionfortrainingsupportvectormachines
_version_	1718373185555005440

DPWSS: differentially private working set selection for training support vector machines

Ejemplares similares