期刊
IEEE INTERNET OF THINGS JOURNAL
卷 8, 期 8, 页码 6178-6186出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/JIOT.2020.3022911
关键词
FedAVG algorithm; federated learning (FL); privacy preservation; secure multiparty computing (SMC)
资金
- Australian Research Council [LP190100676]
- Science Research and Development Program of Jilin Province in China [20200401076GX, 20190303021SF]
- Natural Science Funds of Jilin Province in China [20190201180JC]
- State Scholarship Fund of CSC in China [201902335003]
- Australian Data61 [CRP c020996]
- Australian Research Council [LP190100676] Funding Source: Australian Research Council
The proposed privacy-preserving FL framework chain-PPFL utilizes single-masking and chained-communication mechanisms to achieve practical privacy preservation with some communication cost, while maintaining the accuracy and convergence speed of the training model. Extensive experiments demonstrate the effectiveness of the scheme in protecting sensitive information of participants in FL.
Federated learning (FL) is a promising new technology in the field of IoT intelligence. However, exchanging model-related data in FL may leak the sensitive information of participants. To address this problem, we propose a novel privacy-preserving FL framework based on an innovative chained secure multiparty computing technique, named chain-PPFL. Our scheme mainly leverages two mechanisms: 1) single-masking mechanism that protects information exchanged between participants and 2) chained-communication mechanism that enables masked information to be transferred between participants with a serial chain frame. We conduct extensive simulation-based experiments using two public data sets (MNIST and CIFAR-100) by comparing both training accuracy and leak defence with other state-of-the-art schemes. We set two data sample distributions (IID and NonIID) and three training models (CNN, MLP, and L-BFGS) in our experiments. The experimental results demonstrate that the chain-PPFL scheme can achieve practical privacy preservation (equivalent to differential privacy with epsilon approaching zero) for FL with some cost of communication and without impairing the accuracy and convergence speed of the training model.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据