TY - GEN
T1 - SFPDML
T2 - 7th International Conference on Mobile Internet Security, MobiSec 2023
AU - Wang, Hongxiao
AU - Jiang, Zoe L.
AU - Zhao, Yanmin
AU - Yiu, Siu Ming
AU - Yang, Peng
AU - Chen, Man
AU - Tan, Zejiu
AU - Jin, Bohan
N1 - Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.
PY - 2024
Y1 - 2024
N2 - In recent years, distributed machine learning has garnered significant attention. However, privacy continues to be an unresolved issue within this field. Multi-key homomorphic encryption over torus (MKTFHE) is one of the promising candidates for addressing this concern. Nevertheless, there may be security risks in the decryption of MKTFHE. Moreover, to our best known, the latest works about MKTFHE only support Boolean operation and linear operation which cannot directly compute the non-linear function like Sigmoid. Therefore, it’s still hard to perform common machine learning such as logistic regression and neural networks in high performance. In this paper, we first discover a possible attack on the existing distributed decryption protocol for MKTFHE and subsequently introduce secret sharing to propose a securer one. Next, we design some tools to implement logistic regression and neural network training in MKTFHE. Comparing the efficiency and accuracy between using Taylor polynomials of Sigmoid and our proposed function as an activation function, the experiments show that the efficiency of our function is 5-10× higher than using Taylor polynomials straightly and keeping a similar accuracy.
AB - In recent years, distributed machine learning has garnered significant attention. However, privacy continues to be an unresolved issue within this field. Multi-key homomorphic encryption over torus (MKTFHE) is one of the promising candidates for addressing this concern. Nevertheless, there may be security risks in the decryption of MKTFHE. Moreover, to our best known, the latest works about MKTFHE only support Boolean operation and linear operation which cannot directly compute the non-linear function like Sigmoid. Therefore, it’s still hard to perform common machine learning such as logistic regression and neural networks in high performance. In this paper, we first discover a possible attack on the existing distributed decryption protocol for MKTFHE and subsequently introduce secret sharing to propose a securer one. Next, we design some tools to implement logistic regression and neural network training in MKTFHE. Comparing the efficiency and accuracy between using Taylor polynomials of Sigmoid and our proposed function as an activation function, the experiments show that the efficiency of our function is 5-10× higher than using Taylor polynomials straightly and keeping a similar accuracy.
KW - Distributed machine learning
KW - Multi-key decryption
KW - Multi-key fully homomorphic encryption
KW - Privacy-preserving machine learning
UR - https://www.scopus.com/pages/publications/85200723972
U2 - 10.1007/978-981-97-4465-7_7
DO - 10.1007/978-981-97-4465-7_7
M3 - 会议稿件
AN - SCOPUS:85200723972
SN - 9789819744640
T3 - Communications in Computer and Information Science
SP - 94
EP - 108
BT - Mobile Internet Security - 7th International Conference, MobiSec 2023, Revised Selected Papers
A2 - You, Ilsun
A2 - Kim, Hwankuk
A2 - Choras, Michal
A2 - Shin, Seonghan
A2 - Astillo, Philip Virgil
PB - Springer Science and Business Media Deutschland GmbH
Y2 - 19 December 2023 through 21 December 2023
ER -