Enhancing poisoning attack mitigation in federated learning through perturbation-defense complementarity on history gradients

Cong Wang, Zhilong Mi, Ziqiao Yin, Binghui Guo