arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2009.08097cs.LGcs.AIstat.ML

An Extension of Fano's Inequality for Characterizing Model Susceptibility to Membership Inference Attacks

University of Texas at San Antonio（德克萨斯大学圣安东尼奥分校）；SRI International（SRI国际公司）；University of Central Florida（中佛罗里达大学）；Oakland University（奥克兰大学）；Air Force Research Laboratory（空军研究实验室）；Oak Ridge National Laboratory（橡树岭国家实验室）；Army Research Laboratory（陆军研究实验室）

Sumit Kumar Jha, Susmit Jha, Rickard Ewetz, Sunny Raj, Alvaro Velasquez, Laura L. Pullum, Ananthram Swami

备注：
9 pages, 3 figures
阅读 PDF
arXiv 页面 ↗HTML 原文 ↗

英文摘要：

Deep neural networks have been shown to be vulnerable to membership inference attacks wherein the attacker aims to detect whether specific input data were used to train the model. These attacks can potentially leak private or proprietary data. We present a new extension of Fano's inequality and employ it to theoretically establish that the probability of success for a membership inference attack on a deep neural network can be bounded using the mutual information between its inputs and its activations. This enables the use of mutual information to measure the susceptibility of a DNN model to membership inference attacks. In our empirical evaluation, we show that the correlation between the mutual information and the susceptibility of the DNN model to membership inference attacks is 0.966, 0.996, and 0.955 for CIFAR-10, SVHN and GTSRB models, respectively.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。

↑