A Seesaw Model Attack Algorithm for Distributed Learning

Kun Yang, Tianyi Luo, Yanjie Dong +1

We investigate the Byzantine attack problem within the context of model training in distributed learning systems. While ensuring the convergence of current model training processes…