Information Leakage Measures for Imperfect Statistical Information: Application to Non-Bayesian Framework

被引:0
作者
Sakib, Shahnewaz Karim [1 ]
Amariucai, George T. [2 ]
Guan, Yong [3 ]
机构
[1] Univ Tennessee Chattanooga, Dept Comp Sci & Engn, Chattanooga, TN 37403 USA
[2] Kansas State Univ, Dept Comp Sci, Manhattan, KS 66502 USA
[3] Iowa State Univ, Dept Elect & Comp Engn, Ames, IA 50011 USA
关键词
Measurement; Privacy; Servers; Bayes methods; Information leakage; Data privacy; Scalability; Machine learning algorithms; Usability; Training; imperfect statistical information; subjective leakage; objective leakage; confidence boost; quantifying privacy; non-Bayesian framework; PRIVACY; PROBABILITY;
D O I
10.1109/TIFS.2024.3516585
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
This paper analyzes the problem of estimating information leakage when the complete statistics of the privacy mechanism are not known, and the only available information consists of several input-output pairs obtained through interaction with the system or through some side channel. Several metrics, such as subjective leakage, objective leakage, and confidence boost, were introduced before for this purpose, but by design only work in a Bayesian framework. However, it is known that Bayesian inference can quickly become intractable if the domains of the involved variables are large. In this paper, we focus on this exact problem and propose a novel approach to perform an estimation of the leakage measures when the true knowledge of the privacy mechanism is beyond the reach of the user for a non-Bayesian framework using machine learning. Initially, we adapt the definition of leakage metrics to a non-Bayesian framework and derive their statistical bounds, and afterward, we evaluate the performance of those metrics via various experiments using Neural Networks, Random Forest Classifiers, and Support Vector Machines. We have also evaluated their performance on an image dataset to demonstrate the versatility of the metrics. Finally, we provide a comparative analysis between our proposed metrics and the metrics of the Bayesian framework.
引用
收藏
页码:1065 / 1080
页数:16
相关论文
共 64 条
[11]   3D Augmented Reality Tangible User Interface using Commodity Hardware [J].
Chamzas, Dimitris ;
Moustakas, Konstantinos .
PROCEEDINGS OF THE 15TH INTERNATIONAL JOINT CONFERENCE ON COMPUTER VISION, IMAGING AND COMPUTER GRAPHICS THEORY AND APPLICATIONS, VOL 1: GRAPP, 2020, :384-391
[12]   Probability of error in information-hiding protocols [J].
Chatzikokolakis, Konstantinos ;
Palamidessi, Catuscia ;
Panangaden, Prakash .
20TH IEEE COMPUTER SECURITY FOUNDATIONS SYMPOSIUM (CSFS20), PROCEEDINGS, 2007, :341-+
[13]  
Chatzikokolakis K, 2010, LECT NOTES COMPUT SC, V6015, P390, DOI 10.1007/978-3-642-12002-2_33
[14]   F-BLEAU: Fast Black-box Leakage Estimation [J].
Cherubin, Giovanni ;
Chatzikokolakis, Konstantinos ;
Palamidessi, Catuscia .
2019 IEEE SYMPOSIUM ON SECURITY AND PRIVACY (SP 2019), 2019, :835-852
[15]   Quantification of integrity [J].
Clarkson, Michael R. ;
Schneider, Fred B. .
MATHEMATICAL STRUCTURES IN COMPUTER SCIENCE, 2015, 25 (02) :207-258
[16]   Quantifying information flow with beliefs [J].
Clarkson, Michael R. ;
Myers, Andrew C. ;
Schneider, Fred B. .
JOURNAL OF COMPUTER SECURITY, 2009, 17 (05) :655-701
[17]  
Czyz P, 2023, Arxiv, DOI [arXiv:2306.11078, DOI 10.48550/ARXIV.2306.11078]
[18]  
Desfontaines Damien, 2020, Proceedings on Privacy Enhancing Technologies, V2020, P288, DOI 10.2478/popets-2020-0028
[19]   Calibrating noise to sensitivity in private data analysis [J].
Dwork, Cynthia ;
McSherry, Frank ;
Nissim, Kobbi ;
Smith, Adam .
THEORY OF CRYPTOGRAPHY, PROCEEDINGS, 2006, 3876 :265-284
[20]  
Fano RM., 1961, AM J PHYS, V29, P793, DOI [10.1119/1.1937609, DOI 10.1119/1.1937609]