Jul 2026
Defending Against Backdoor Attacks via Alignment Checking in Model-Contrastive Federated Learning
FedDAB, a two-phase method that combines local contrastive regularization with alignment checking, to defend against backdoor attacks, is proposed and theoretically proves FedDAB's robustness with a convergence rate of $\mathcal{O}(1/T)$.
Hongliang Zhang, Zhongyuan Yu, Guijuan Wang et al.
· arXiv.org · 0 citations