《计算机应用》唯一官方网站 ›› 2022, Vol. 42 ›› Issue (12): 3792-3800.DOI: 10.11772/j.issn.1001-9081.2021101775

• 网络空间安全 • 上一篇    

支持隐私保护训练的高效同态神经网络

钟洋1,2, 毕仁万1,2, 颜西山1, 应作斌3, 熊金波1,2()   

  1. 1.福建师范大学 计算机与网络空间安全学院, 福州 350117
    2.福建省网络安全与密码技术重点实验室(福建师范大学), 福州 350007
    3.澳门城市大学 数据科学学院, 澳门 999078
  • 收稿日期:2021-10-15 修回日期:2022-01-18 接受日期:2022-01-24 发布日期:2022-03-04 出版日期:2022-12-10
  • 通讯作者: 熊金波
  • 作者简介:钟洋(1995—),男,湖南湘西人,硕士研究生,CCF会员,主要研究方向:安全深度学习
    毕仁万(1996—),男,湖南常德人,博士研究生,CCF会员,主要研究方向:安全深度学习、安全多方计算
    颜西山(1979—),男,江西吉安人,博士研究生,主要研究方向:信息安全
    应作斌(1982—),男,安徽芜湖人,助理教授,博士,主要研究方向:云计算安全
  • 基金资助:
    国家自然科学基金资助项目(61872088)

Efficient homomorphic neural network supporting privacy-preserving training

Yang ZHONG1,2, Renwan BI1,2, Xishan YAN1, Zuobin YING3, Jinbo XIONG1,2()   

  1. 1.College of Computer and Cyber Security,Fujian Normal University,Fuzhou Fujian 350117,China
    2.Fujian Provincial Key Lab of Network Security and Cryptology (Fujian Normal University),Fuzhou Fujian 350007,China
    3.Faculty of Data Science,City University of Macau,Macau 999078,China
  • Received:2021-10-15 Revised:2022-01-18 Accepted:2022-01-24 Online:2022-03-04 Published:2022-12-10
  • Contact: Jinbo XIONG
  • About author:ZHONG Yang, born in 1995, M. S. candidate. His research interests include secure deep learning.
    BI Renwan, born in 1996, Ph. D. candidate. His research interests include secure deep learning, secure multi-party computing.
    YAN Xishan, born in 1979, Ph. D. candidate. His research interests include information security.
    YING Zuobin, born in 1982, Ph. D., assistant professor. His research interests include security for cloud computing.
  • Supported by:
    National Natural Science Foundation of China(61872088)

摘要:

针对基于同态加密的隐私保护神经网络中存在的计算效率低和精度不足问题,提出一种三方协作下支持隐私保护训练的高效同态神经网络(HNN)。首先,为降低同态加密中密文乘密文运算产生的计算开销,结合秘密共享思想设计了一种安全快速的乘法协议,将密文乘密文运算转换为复杂度较低的明文乘密文运算;其次,为避免构建HNN时产生的密文多项式多轮迭代,并提高非线性计算精度,研究了一种安全的非线性计算方法,从而对添加随机掩码的混淆明文消息执行相应的非线性算子;最后,对所设计协议的安全性、正确性及效率进行了理论分析,并对HNN的有效性及优越性进行了实验验证。实验结果表明,相较于双服务器方案PPML,HNN的训练速度提高了18.9倍,模型精度提高了1.4个百分点。

关键词: 同态加密, 隐私保护, 神经网络, 模型训练, 非线性计算

Abstract:

Aiming at the problems of low computational efficiency and insufficient accuracy in the privacy-preserving neural network based on homomorphic encryption, an efficient Homomorphic Neural Network (HNN) under three-party collaborative supporting privacy-preserving training was proposed. Firstly, in order to reduce the computational cost of ciphertext-ciphertext multiplication in homomorphic encryption, the idea of secret sharing was combined to design a secure fast multiplication protocol to convert the ciphertext-ciphertext multiplication into plaintext-ciphertext multiplication with low complexity. Then, in order to avoid multiple iterations of ciphertext polynomials generated during the construction of HNN and improve the nonlinear calculation accuracy, a secure nonlinear calculation method was studied, which executed the corresponding nonlinear operator for the confused plaintext message with random mask. Finally, the security, correctness and efficiency of the proposed protocols were analyzed theoretically, and the effectiveness and superiority of HNN were verified by experiments. Experimental results show that compared with the dual server scheme PPML (Privacy Protection Machine Learning), HNN has the training efficiency improved by 18.9 times and the model accuracy improved by 1.4 percentage points.

Key words: homomorphic encryption, privacy-preserving, neural network, model training, nonlinear calculation

中图分类号: