支持层级相关性传播的差分隐私分类算法研究

侯小军; 李泽华; 李泽堃

Chinese

您当前的位置：

首页 >

文章列表页 >

支持层级相关性传播的差分隐私分类算法研究

更新时间：2026-09-04

- 支持层级相关性传播的差分隐私分类算法研究
- Research on differentially private classification algorithm based on layer-wise relevance propagation
- 微电子学与计算机 2021年38卷第5期页码：48-53
- 作者机构：
  
  1. 西安理工大学机构信息化管理处,陕西,西安,710048
  2. 陕西师范大学计算机科学学院,陕西,西安,710062
- 作者简介：
- 基金信息：
  
  西安市科技计划项目 2019216914GXRC005CG006-GXYD5.2中央高校基本业务费项目 GK201906009赛尔网络下一代互联网技术创新项目 NGII20190704
- DOI：
  中图分类号：
- 纸质出版：2021
- 稿件说明：
移动端阅览
侯小军, 李泽华, 李泽堃. 支持层级相关性传播的差分隐私分类算法研究[J]. 微电子学与计算机, 2021, 38(5): 48-53.

HOU Xiao-jun, LI Ze-hua, LI Ze-kun. Research on differentially private classification algorithm based on layer-wise relevance propagation[J]. 2021, 38(5): 48-53.
侯小军, 李泽华, 李泽堃. 支持层级相关性传播的差分隐私分类算法研究[J]. 微电子学与计算机, 2021, 38(5): 48-53. DOI：

HOU Xiao-jun, LI Ze-hua, LI Ze-kun. Research on differentially private classification algorithm based on layer-wise relevance propagation[J]. 2021, 38(5): 48-53. DOI：

摘要

为了防止攻击者在深度学习图像分类过程中还原训练集数据并保护输入图像数据，提出一种基于层级相关性传播的差分隐私分类算法.该算法首先采用层级相关性传播模型量化图像的特征相关性，然后利用相关性自适应地向损失函数添加噪声并利用Adam算法进行模型优化，最后依据相关性分配隐私预算并构造差分隐私变换层以扰动输入数据.实验结果表明，该算法在实现隐私保护的同时，能够保证较高的分类准确率.

Abstract

To prevent attackers from restoring the training dataset and to protect the input image during the application process of deep learning image classification model

a differential privacy classification algorithm based on LRP (Layer-wise Relevance Propagation) is proposed in the paper. The relevance between image features is firstly quantified according to LRP in the proposed algorithm

then adaptive noise is added to the loss function based on feature relevance and Adam mechanism is used for model optimization. Finally

a differential private transform layer is constructed to perturb the input image and privacy budget is assigned according to feature relevance during model application stage. Experimental results show that the proposed algorithm achieves high classification accuracy in condition of privacy preservation.

关键词

Keywords

references

PHAN N H, WANG Y, WU X T, et al. Differential privacy preservation for deep auto-encoders: an application of human behavior prediction[C]//Proceedings of the 30th AAAI Conference on Artificial Intelligence. Phoenix: AAAI, 2016: 1309-1316.

ZHAO J W, CHEN Y F, ZHANG W. Differential privacy preservation in deep learning: challenges, opportunities and solutions[J]. IEEE Access, 2019(7): 48901-48911. DOI: 10.1109/ACCESS.2019.2909559.

ABADI M, CHU A, GOODFELLOW I, et al. Deep learning with differential privacy[C]//Proceedings of 2016 ACM SIGSAC Conference on Computer and Communications Security. Vienna: ACM, 2016: 308-318. DOI: 10.1145/2976749.2978318.

FREDRIKSON M, LANTZ E, JHA S, et al. Privacy in pharmacogenetics: an end-to-end case study of personalized warfarin dosing[C]//Proceedings of the 23rd USENIX Conference on Security Symposium. San Diego: USENIX Association, 2014. DOI: 10.5555/2671225.2671227.

HAEBERLEN A, PIERCE B C, NARAYAN A. Differential privacy under fire[C]//Proceedings of the 20th USENIX Conference on Security. San Francisco: USENIX Association, 2011: 33. DOI: 10.5555/2028067.2028100.

MCSHERRY F D. Privacy integrated queries: an extensible platform for privacy-preserving data analysis[C]//Proceedings of 2009 ACM SIGMOD International Conference on Management of Data. Providence: ACM, 2009: 19-30. DOI: 10.1145/1559845.1559850.

XIE L Y, LIN K X, WANG S, et al. Differentially private generative adversarial network[J]. arXiv: 1802.06739, 2018.

PHAN N H, WU X T, HU H, et al. Adaptive Laplace mechanism: differential privacy preservation in deep learning[C]//Proceedings of 2017 IEEE International Conference on Data Mining. New Orleans: IEEE, 2017: 385-394. DOI: 10.1109/ICDM.2017.48.

DWORK C. Differential privacy[C]//Proceedings of the 33rd International Colloquium on Automata, Languages, and Programming. Venice, Italy: Springer, 2006. DOI: 10.1007/11787006_1.

BACH S, BINDER A, MONTAVON G, et al. On pixel-wise explanations for non-linear classifier decisions by layer-wise relevance propagation[J]. PLoS One, 2015, 10(7): e0130140. DOI: 10.1371/journal.pone.0130140.

PHAN N H, WU X T, DOU D J. Preserving differential privacy in convolutional deep belief networks[J]. Machine Learning, 2017, 106(9-10): 1681-1704. DOI: 10.1007/s10994-017-5656-2.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于计算机视觉的电子元器件表面缺陷检测综述

基于差异增强与边缘感知的耕地变化检测方法研究

改进RT-DETR的无人机小目标检测算法

多模态双注意力机制点云语义分割方法

基于注意力机制改进的DeepLabV3+遥感图像分割算法