Efficient Differentially Private Secure Aggregation for Federated Learning via Hardness of Learning with Errors
Efficient Differentially Private Secure Aggregation for Federated Learning via Hardness of Learning with Errors
复制标题
DOI：
--
复制
发表时间：
2021-12
期刊：
影响因子：
--
通讯作者：
Timothy Stevens;C. Skalka;C. Vincent;J. Ring;Samuel Clark;Joseph P. Near
中科院分区：
文献类型：
--
作者：
Timothy Stevens;C. Skalka;C. Vincent;J. Ring;Samuel Clark;Joseph P. Near
联邦机器学习利用边缘计算从网络用户数据开发模型，但联邦学习中的隐私仍然是一个重大挑战。人们已经提出使用差异隐私的技术来解决这个问题，但也带来了自己的挑战——许多技术需要可信的第三方，否则会增加太多噪音来生成有用的模型。使用多方计算的 emph{安全聚合} 的最新进展消除了对第三方的需求，但计算成本高昂，尤其是在规模上。我们提出了一种新的联合学习协议，该协议利用了一种基于错误学习技术的新颖的差分隐私、恶意安全聚合协议。我们的协议优于当前最先进的技术，并且实证结果表明它可以扩展到大量参与方，并且对于任何差分私有联邦学习方案都具有最佳准确性。
Federated machine learning leverages edge computing to develop models from network user data, but privacy in federated learning remains a major challenge. Techniques using differential privacy have been proposed to address this, but bring their own challenges -- many require a trusted third party or else add too much noise to produce useful models. Recent advances in \emph{secure aggregation} using multiparty computation eliminate the need for a third party, but are computationally expensive especially at scale. We present a new federated learning protocol that leverages a novel differentially private, malicious secure aggregation protocol based on techniques from Learning With Errors. Our protocol outperforms current state-of-the art techniques, and empirical results show that it scales to a large number of parties, with optimal accuracy for any differentially private federated learning scheme.