AI RESEARCH
Convergence of Byzantine-Resilient Gradient Tracking via Probabilistic Edge Dropout
arXiv CS.LG
•
ArXi:2604.00449v1 Announce Type: new We study distributed optimization over networks with Byzantine agents that may send arbitrary adversarial messages. We propose \emph{Gradient Tracking with Probabilistic Edge Dropout} (GT-PD), a stochastic gradient tracking method that preserves the convergence properties of gradient tracking under adversarial communication.