TY - GEN
T1 - Speed-up of Data Analysis with Kernel Trick in Encrypted Domain
AU - Yoo, Joon Soo
AU - Song, Baek Kyung
AU - Ahn, Tae Min
AU - Heo, Jiwon
AU - Yoon, Ji Won
N1 - Publisher Copyright:
Copyright © 2025 held by the owner/author(s).
PY - 2025/5/14
Y1 - 2025/5/14
N2 - Fully Homomorphic Encryption (FHE) is pivotal for secure computation on encrypted data, crucial in privacy-preserving data analysis. However, efficiently processing high-dimensional data in FHE, especially for machine learning and statistical (ML/STAT) algorithms, poses a challenge. In this paper, we present an effective acceleration method using the kernel method for FHE schemes, enhancing time performance in ML/STAT algorithms within encrypted domains. This technique, independent of underlying FHE mechanisms and complementing existing optimizations, notably reduces costly FHE multiplications, offering near-constant time complexity relative to data dimension. Aimed at accessibility, this method is tailored for data scientists and developers with limited cryptography background, facilitating advanced data analysis in secure environments.
AB - Fully Homomorphic Encryption (FHE) is pivotal for secure computation on encrypted data, crucial in privacy-preserving data analysis. However, efficiently processing high-dimensional data in FHE, especially for machine learning and statistical (ML/STAT) algorithms, poses a challenge. In this paper, we present an effective acceleration method using the kernel method for FHE schemes, enhancing time performance in ML/STAT algorithms within encrypted domains. This technique, independent of underlying FHE mechanisms and complementing existing optimizations, notably reduces costly FHE multiplications, offering near-constant time complexity relative to data dimension. Aimed at accessibility, this method is tailored for data scientists and developers with limited cryptography background, facilitating advanced data analysis in secure environments.
KW - fully homomorphic encryption
KW - high-dimensional data analysis
KW - kernel method
KW - privacy-preserving machine learning
UR - https://www.scopus.com/pages/publications/105006455278
U2 - 10.1145/3672608.3707713
DO - 10.1145/3672608.3707713
M3 - Conference paper
AN - SCOPUS:105006455278
T3 - Proceedings of the ACM Symposium on Applied Computing
SP - 1055
EP - 1064
BT - 40th Annual ACM Symposium on Applied Computing, SAC 2025
PB - Association for Computing Machinery
T2 - 40th Annual ACM Symposium on Applied Computing, SAC 2025
Y2 - 31 March 2025 through 4 April 2025
ER -