Zheng Yang, Ke Gu*, Yiming Zuo
CMC-Computers, Materials & Continua, Vol.79, No.2, pp. 2813-2831, 2024, DOI:10.32604/cmc.2024.050025
Abstract Federated learning is widely used to solve the problem of data decentralization and can provide privacy protection for data owners. However, since multiple participants are required in federated learning, this allows attackers to compromise. Byzantine attacks pose great threats to federated learning. Byzantine attackers upload maliciously created local models to the server to affect the prediction performance and training speed of the global model. To defend against Byzantine attacks, we propose a Byzantine robust federated learning scheme based on backdoor triggers. In our scheme, backdoor triggers are embedded into benign data samples, and then malicious More >
Login
Register
