arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2102.10639cs.ITeess.SPmath.IT
收藏

利用固有硬件缺陷的隐私保护无线联邦学习

Privacy-Preserving Wireless Federated Learning Exploiting Inherent Hardware Impairments

Sina Rezaei Aghdam, Ehsan Amid, Marija Furdek, Alexandre Graell i Amat

摘要:

我们考虑了一个无线联邦学习系统,其中多个数据持有者边缘设备通过与诚实但好奇的参数服务器共享参数更新来协作训练全局模型。我们证明了固有的硬件引起的失真扰动的边缘设备的模型更新可以被利用作为一种隐私保护机制。特别是,我们的失真模型为功率依赖的加性高斯噪声,并提出了一个功率分配策略,提供隐私保证的框架内的差分隐私。我们进行数值实验,以评估所提出的功率分配方案的性能在不同程度的硬件损伤。

英文摘要:

We consider a wireless federated learning system where multiple data holder edge devices collaborate to train a global model via sharing their parameter updates with an honest-but-curious parameter server. We demonstrate that the inherent hardware-induced distortion perturbing the model updates of the edge devices can be exploited as a privacy-preserving mechanism. In particular, we model the distortion as power-dependent additive Gaussian noise and present a power allocation strategy that provides privacy guarantees within the framework of differential privacy. We conduct numerical experiments to evaluate the performance of the proposed power allocation scheme under different levels of hardware impairments.

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。

↑