arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1812.02863cs.CRcs.CVcs.LG
收藏

隐私划分:在深度学习推理阶段保护用户数据

Privacy Partitioning: Protecting User Data During the Deep Learning Inference Phase

University of Virginia(弗吉尼亚大学);Carnegie Mellon University(卡内基梅隆大学);Google Brain(谷歌大脑)

Jianfeng Chi, Emmanuel Owusu, Xuwang Yin, Tong Yu, William Chan, Patrick Tague, Yuan Tian

摘要:

我们提出了一种基于二分拓扑威胁建模和交互式对抗深度网络构建的深度学习推理阶段的数据保护实用方法。我们将这种方法称为\emph{Privacy Partitioning}。在提出的框架中,我们拆分了机器学习模型,并将几层部署到用户的本地设备中,其余层部署到远程服务器中。我们提出了一种方法来保护用户的数据在推理阶段,同时仍然实现良好的分类精度。我们对三个计算机视觉任务的基准数据集进行了实验评估。实验结果表明,这种方法可以用来显着削弱攻击者访问最先进的深度网络的中间状态来学习网络的隐私敏感输入的能力。例如,我们证明了我们的方法可以防止攻击者从人脸图像数据集中推断出性别等私有属性,而不会牺牲原始机器学习任务(如人脸识别)的分类准确性。

英文摘要:

We present a practical method for protecting data during the inference phase of deep learning based on bipartite topology threat modeling and an interactive adversarial deep network construction. We term this approach \emph{Privacy Partitioning}. In the proposed framework, we split the machine learning models and deploy a few layers into users' local devices, and the rest of the layers into a remote server. We propose an approach to protect user's data during the inference phase, while still achieve good classification accuracy. We conduct an experimental evaluation of this approach on benchmark datasets of three computer vision tasks. The experimental results indicate that this approach can be used to significantly attenuate the capacity for an adversary with access to the state-of-the-art deep network's intermediate states to learn privacy-sensitive inputs to the network. For example, we demonstrate that our approach can prevent attackers from inferring the private attributes such as gender from the Face image dataset without sacrificing the classification accuracy of the original machine learning task such as Face Identification.

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。

↑