arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1709.06161cs.LG
收藏

PrivyNet:一个灵活的隐私保护深度神经网络训练框架

PrivyNet: A Flexible Framework for Privacy-Preserving Deep Neural Network Training

University of Texas at Austin(德克萨斯大学奥斯汀分校);Arm Inc.(Arm 公司)

Meng Li, Liangzhen Lai, Naveen Suda, Vikas Chandra, David Z. Pan

摘要:

海量数据存在于用户本地平台之间,由于计算和存储资源的限制,这些平台通常无法支持深度神经网络(DNN)训练。基于云的培训方案提供有益的服务,但由于过度收集用户数据而遭受潜在的隐私风险。为了实现基于云的DNN训练,同时保护数据隐私,我们建议利用数据的中间表示,这是通过拆分DNN并将其单独部署到本地平台和云上来实现的。局部神经网络(NN)用于生成特征表示。为了避免局部训练和保护数据隐私,局部NN来自预训练的NN。然后,基于所提取的目标学习任务的中间表示来训练云NN。我们验证DNN分裂的想法,通过表征的依赖性的隐私损失和分类精度的局部NN拓扑结构的卷积神经网络(CNN)为基础的图像分类任务。基于特征,我们进一步提出了PrivyNet来确定局部NN拓扑结构,该拓扑结构在隐私损失、局部计算和存储的约束下优化了目标学习任务的准确性。PrivyNet的效率和有效性与CIFAR-10数据集证明。

英文摘要:

Massive data exist among user local platforms that usually cannot support deep neural network (DNN) training due to computation and storage resource constraints. Cloud-based training schemes provide beneficial services but suffer from potential privacy risks due to excessive user data collection. To enable cloud-based DNN training while protecting the data privacy simultaneously, we propose to leverage the intermediate representations of the data, which is achieved by splitting the DNNs and deploying them separately onto local platforms and the cloud. The local neural network (NN) is used to generate the feature representations. To avoid local training and protect data privacy, the local NN is derived from pre-trained NNs. The cloud NN is then trained based on the extracted intermediate representations for the target learning task. We validate the idea of DNN splitting by characterizing the dependency of privacy loss and classification accuracy on the local NN topology for a convolutional NN (CNN) based image classification task. Based on the characterization, we further propose PrivyNet to determine the local NN topology, which optimizes the accuracy of the target learning task under the constraints on privacy loss, local computation, and storage. The efficiency and effectiveness of PrivyNet are demonstrated with the CIFAR-10 dataset.

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。

↑