Adaptive Laplace Mechanism: Differential Privacy Preservation in Deep Learning

被引:114
|
作者
Phan, NhatHai [1 ]
Wu, Xintao [2 ]
Hu, Han [1 ]
Dou, Dejing [3 ]
机构
[1] New Jersey Inst Technol, Newark, NJ 07102 USA
[2] Univ Arkansas, Fayetteville, AR 72701 USA
[3] Univ Oregon, Eugene, OR 97403 USA
关键词
D O I
10.1109/ICDM.2017.48
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
In this paper, we focus on developing a novel mechanism to preserve differential privacy in deep neural networks, such that: (1) The privacy budget consumption is totally independent of the number of training steps; (2) It has the ability to adaptively inject noise into features based on the contribution of each to the output; and (3) It could be applied in a variety of different deep neural networks. To achieve this, we figure out a way to perturb affine transformations of neurons, and loss functions used in deep neural networks. In addition, our mechanism intentionally adds "more noise" into features which are "less relevant" to the model output, and vice-versa. Our theoretical analysis further derives the sensitivities and error bounds of our mechanism. Rigorous experiments conducted on MNIST and CIFAR-10 datasets show that our mechanism is highly effective and outperforms existing solutions.
引用
下载
收藏
页码:385 / 394
页数:10
相关论文
共 50 条
  • [21] Differential Privacy Preservation in Adaptive K-Nets Clustering
    Liu, Xiaohong
    Cai, Hanbo
    Li, De
    Li, Xianxian
    Wang, Jinyan
    2021 IEEE 20TH INTERNATIONAL CONFERENCE ON TRUST, SECURITY AND PRIVACY IN COMPUTING AND COMMUNICATIONS (TRUSTCOM 2021), 2021, : 405 - 412
  • [22] Grafting Laplace and Gaussian Distributions: A New Noise Mechanism for Differential Privacy
    Muthukrishnan, Gokularam
    Kalyani, Sheetal
    IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY, 2023, 18 : 5359 - 5374
  • [23] Research on the Deep Learning Method Based on Data Feature Relevance and Adaptive Differential Privacy
    Kang, Hai-Yan
    Wang, Xiao-Shi
    Tien Tzu Hsueh Pao/Acta Electronica Sinica, 2024, 52 (06): : 1963 - 1976
  • [24] Medical imaging deep learning with differential privacy
    Ziller, Alexander
    Usynin, Dmitrii
    Braren, Rickmer
    Makowski, Marcus
    Rueckert, Daniel
    Kaissis, Georgios
    SCIENTIFIC REPORTS, 2021, 11 (01)
  • [25] Towards Decentralized Deep Learning with Differential Privacy
    Cheng, Hsin-Pai
    Yu, Patrick
    Hu, Haojing
    Zawad, Syed
    Yan, Feng
    Li, Shiyu
    Li, Hai
    Chen, Yiran
    CLOUD COMPUTING - CLOUD 2019, 2019, 11513 : 130 - 145
  • [26] Differential Privacy Preserving Deep Learning in Healthcare
    Wu, Xintao
    2017 IEEE INTERNATIONAL CONFERENCE ON BIOINFORMATICS AND BIOMEDICINE (BIBM), 2017, : 8 - 8
  • [27] Differential privacy in deep learning: A literature survey
    Pan, Ke
    Ong, Yew-Soon
    Gong, Maoguo
    Li, Hui
    Qin, A. K.
    Gao, Yuan
    NEUROCOMPUTING, 2024, 589
  • [28] Differential Privacy for Deep and Federated Learning: A Survey
    El Ouadrhiri, Ahmed
    Abdelhadi, Ahmed
    IEEE ACCESS, 2022, 10 : 22359 - 22380
  • [29] Medical imaging deep learning with differential privacy
    Alexander Ziller
    Dmitrii Usynin
    Rickmer Braren
    Marcus Makowski
    Daniel Rueckert
    Georgios Kaissis
    Scientific Reports, 11
  • [30] Trajectory time prediction and dataset publishing mechanism based on deep learning and differential privacy
    Li, Dongping
    Shen, Shikai
    Yang, Yingchun
    He, Jun
    Shen, Haoru
    JOURNAL OF INTELLIGENT & FUZZY SYSTEMS, 2023, 45 (01) : 783 - 795