Privacy-Preserving Student Learning with Differentially Private Data-Free Distillation

被引:2
|
作者
Liu, Bochao [1 ,2 ]
Lu, Jianghu [1 ,2 ]
Wang, Pengju [1 ,2 ]
Zhang, Junjie [3 ]
Zeng, Dan [3 ]
Qian, Zhenxing [4 ]
Ge, Shiming [1 ,2 ]
机构
[1] Chinese Acad Sci, Inst Informat Engn, Beijing 100095, Peoples R China
[2] Univ Chinese Acad Sci, Sch Cyber Secur, Beijing 100049, Peoples R China
[3] Shanghai Univ, Sch Commun & Informat Engn, Shanghai 200444, Peoples R China
[4] Fudan Univ, Sch Comp Sci, Shanghai 200433, Peoples R China
来源
2022 IEEE 24TH INTERNATIONAL WORKSHOP ON MULTIMEDIA SIGNAL PROCESSING (MMSP) | 2022年
基金
北京市自然科学基金;
关键词
differential privacy; teacher-student learning; knowledge distillation;
D O I
10.1109/MMSP55362.2022.9950001
中图分类号
TP31 [计算机软件];
学科分类号
081202 ; 0835 ;
摘要
Deep learning models can achieve high inference accuracy by extracting rich knowledge from massive well-annotated data, but may pose the risk of data privacy leakage in practical deployment. In this paper, we present an effective teacher-student learning approach to train privacy-preserving deep learning models via differentially private data-free distillation. The main idea is generating synthetic data to learn a student that can mimic the ability of a teacher well-trained on private data. In the approach, a generator is first pretrained in a data-free manner by incorporating the teacher as a fixed discriminator. With the generator, massive synthetic data can be generated for model training without exposing data privacy. Then, the synthetic data is fed into the teacher to generate private labels. Towards this end, we propose a label differential privacy algorithm termed selective randomized response to protect the label information. Finally, a student is trained on the synthetic data with the supervision of private labels. In this way, both data privacy and label privacy are well protected in a unified framework, leading to privacy-preserving models. Extensive experiments and analysis clearly demonstrate the effectiveness of our approach.
引用
收藏
页数:6
相关论文
共 50 条
  • [41] Privacy-preserving federated learning on lattice quantization
    Zhang, Lingjie
    Zhang, Hai
    INTERNATIONAL JOURNAL OF WAVELETS MULTIRESOLUTION AND INFORMATION PROCESSING, 2023, 21 (06)
  • [42] Privacy-Preserving News Recommendation Model Learning
    Qi, Tao
    Wu, Fangzhao
    Wu, Chuhan
    Huang, Yongfeng
    Xie, Xing
    FINDINGS OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS, EMNLP 2020, 2020, : 1423 - 1432
  • [43] Achieving Consensus in Privacy-Preserving Decentralized Learning
    Xiang, Liyao
    Wang, Lingdong
    Wang, Shufan
    Li, Baochun
    2020 IEEE 40TH INTERNATIONAL CONFERENCE ON DISTRIBUTED COMPUTING SYSTEMS (ICDCS), 2020, : 899 - 909
  • [44] Privacy-Preserving Cost-Sensitive Learning
    Yang, Yi
    Huang, Shuai
    Huang, Wei
    Chang, Xiangyu
    IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2021, 32 (05) : 2105 - 2116
  • [45] SecDM: privacy-preserving data outsourcing framework with differential privacy
    Dagher, Gaby G.
    Fung, Benjamin C. M.
    Mohammed, Noman
    Clark, Jeremy
    KNOWLEDGE AND INFORMATION SYSTEMS, 2020, 62 (05) : 1923 - 1960
  • [46] A Privacy-Preserving Health Data Aggregation Scheme
    Liu, Yining
    Liu, Gao
    Cheng, Chi
    Xia, Zhe
    Shen, Jian
    KSII TRANSACTIONS ON INTERNET AND INFORMATION SYSTEMS, 2016, 10 (08): : 3852 - 3864
  • [47] Privacy-Preserving Data Mining for Smart Manufacturing
    Hu, Qianyu
    Chen, Ruimin
    Yang, Hui
    Kumara, Soundar
    SMART AND SUSTAINABLE MANUFACTURING SYSTEMS, 2020, 4 (02): : 99 - 120
  • [48] Privacy-Preserving Personal Sensitive Data in Crowdsourcing
    Xu, Ke
    Han, Kai
    Ye, Hang
    Gao, Feng
    Xu, Chaoting
    WIRELESS ALGORITHMS, SYSTEMS, AND APPLICATIONS (WASA 2018), 2018, 10874 : 509 - 520
  • [49] Privacy-Preserving Data Publishing in Process Mining
    Rafiei, Majid
    van der Aalst, Wil M. P.
    BUSINESS PROCESS MANAGEMENT FORUM, BPM FORUM 2020, 2020, 392 : 122 - 138
  • [50] Characterizing and Optimizing Differentially-Private Techniques for High-Utility, Privacy-Preserving Internet-of-Vehicles
    Duan, Yicun
    Liu, Junyu
    Ming, Xiaoxing
    Jin, Wangkai
    Song, Zilin
    Peng, Xiangjun
    HCI IN MOBILITY, TRANSPORT, AND AUTOMOTIVE SYSTEMS, MOBITAS 2023, PT I, 2023, 14048 : 31 - 50