Training Support Vector Machines with privacy-protected data
作者：
Highlights：
• Multiple-key encrypted machine learning scenario.
• Standard authorization protocol (OAuth 2.0) to get access to encrypted data.
• A minimal set of outsourced operations to optimize the encryption/decryption hardware (CryptoProcessor).
• Semiparametric SVM scheme that avoids the use of private instances as part of the model.
• Analysis of the SVMs performance under thenite-precission conditions required by cryptosystems.
摘要
•Multiple-key encrypted machine learning scenario.•Standard authorization protocol (OAuth 2.0) to get access to encrypted data.•A minimal set of outsourced operations to optimize the encryption/decryption hardware (CryptoProcessor).•Semiparametric SVM scheme that avoids the use of private instances as part of the model.•Analysis of the SVMs performance under thenite-precission conditions required by cryptosystems.
论文关键词：Machine learning,Privacy protection,Homomorphic encryption,Support Vector Machines
论文评审过程：Received 31 March 2017, Revised 12 May 2017, Accepted 7 June 2017, Available online 28 June 2017, Version of Record 8 July 2017.
论文官网地址：https://doi.org/10.1016/j.patcog.2017.06.016
