Multi-Agent Adversarial Training Using Diffusion Learning
Ying Shan Cao, Elsa Rizk, Stefan Vlaski, Ali H. Sayed · 2023
This work focuses on adversarial learning over graphs. We propose a general adversarial training framework for multi-agent systems using diffusion learning. We analyze the convergence properties of the proposed scheme for convex optimization problems, and illustrate its enhanced robustness to adversarial attacks.