TY - GEN
T1 - HomoPAI
T2 - 36th IEEE International Conference on Data Engineering, ICDE 2020
AU - Li, Qifei
AU - Huang, Zhicong
AU - Lu, Wen Jie
AU - Hong, Cheng
AU - Qu, Hunter
AU - He, Hui
AU - Zhang, Weizhe
N1 - Publisher Copyright:
© 2020 IEEE.
PY - 2020/4
Y1 - 2020/4
N2 - Homomorphic Encryption (HE) allows encrypted data to be processed without decryption, which could maximize the protection of user privacy without affecting the data utility. Thanks to strides made by cryptographers in the past few years, the efficiency of HE has been drastically improved, and machine learning on homomorphically encrypted data has become possible. Several works have explored machine learning based on HE, but most of them are restricted to the outsourced scenario, where all the data comes from a single data owner. We propose HomoPAI, an HE-based secure collaborative machine learning system, enabling a more promising scenario, where data from multiple data owners could be securely processed. Moreover, we integrate our system with the popular MPI framework to achieve parallel HE computations. Experiments show that our system can train a logistic regression model on millions of homomorphically encrypted data in less than two minutes.
AB - Homomorphic Encryption (HE) allows encrypted data to be processed without decryption, which could maximize the protection of user privacy without affecting the data utility. Thanks to strides made by cryptographers in the past few years, the efficiency of HE has been drastically improved, and machine learning on homomorphically encrypted data has become possible. Several works have explored machine learning based on HE, but most of them are restricted to the outsourced scenario, where all the data comes from a single data owner. We propose HomoPAI, an HE-based secure collaborative machine learning system, enabling a more promising scenario, where data from multiple data owners could be securely processed. Moreover, we integrate our system with the popular MPI framework to achieve parallel HE computations. Experiments show that our system can train a logistic regression model on millions of homomorphically encrypted data in less than two minutes.
KW - Homomorphic encryption
KW - Machine learning
UR - https://www.scopus.com/pages/publications/85085863076
U2 - 10.1109/ICDE48307.2020.00152
DO - 10.1109/ICDE48307.2020.00152
M3 - 会议稿件
AN - SCOPUS:85085863076
T3 - Proceedings - International Conference on Data Engineering
SP - 1713
EP - 1717
BT - Proceedings - 2020 IEEE 36th International Conference on Data Engineering, ICDE 2020
PB - IEEE Computer Society
Y2 - 20 April 2020 through 24 April 2020
ER -