Network-Level Adversaries in Federated Learning
Network-Level Adversaries in Federated Learning
复制标题
DOI：
10.1109/cns56114.2022.9947237
复制
发表时间：
2022-08
期刊：
影响因子：
--
通讯作者：
Giorgio Severi;Matthew Jagielski;Gokberk Yar;Yuxuan Wang;Alina Oprea;C. Nita-Rotaru
中科院分区：
文献类型：
--
作者：
Giorgio Severi;Matthew Jagielski;Gokberk Yar;Yuxuan Wang;Alina Oprea;C. Nita-Rotaru
联邦学习是一种流行的策略，用于在分布式敏感数据上训练模型，同时保护数据隐私。先前的工作确定了联邦学习协议上的一系列安全威胁，这些威胁会破坏数据或模型。然而，联邦学习是一个网络系统，其中客户端和服务器之间的通信对学习任务的性能起着至关重要的作用。我们强调了通信如何在联邦学习中引入另一个漏洞面，并研究了网络级对手对训练联邦学习模型的影响。我们展示了攻击者从精心选择的客户端丢弃网络流量可以显著降低目标人群的模型准确性。此外，我们还表明，少数客户的协同投毒活动可以放大投放攻击。最后，我们开发了一个服务器端防御，通过识别和上采样可能对目标准确性做出积极贡献的客户端来减轻攻击的影响。我们在三个数据集上全面评估了我们的攻击和防御，假设加密的通信通道和攻击者具有部分网络可见性。
Federated learning is a popular strategy for training models on distributed, sensitive data, while preserving data privacy. Prior work identified a range of security threats on federated learning protocols that poison the data or the model. However, federated learning is a networked system where the communication between clients and server plays a critical role for the learning task performance. We highlight how communication introduces another vulnerability surface in federated learning and study the impact of network-level adversaries on training federated learning models. We show that attackers dropping the network traffic from carefully selected clients can significantly decrease model accuracy on a target population. Moreover, we show that a coordinated poisoning campaign from a few clients can amplify the dropping attacks. Finally, we develop a server-side defense which mitigates the impact of our attacks by identifying and up-sampling clients likely to positively contribute towards target accuracy. We comprehensively evaluate our attacks and defenses on three datasets, assuming encrypted communication channels and attackers with partial visibility of the network.