Multi-Agent Adversarial Training Using Diffusion Learning

2023/03/03 by Ying Cao, Cao, Ying, Elsa Rizk +5
Computer Science · Social Sciences · #Adversarial Robustness in Machine Learning #Artificial Intelligence (cs.AI) #FOS: Computer and information sciences #Machine Learning (cs.LG) #Multiagent Systems (cs.MA) #Terrorism, Counterterrorism, and Political Violence

paper · pdf · doi:10.48550/arxiv.2303.01936

openalex publication_date 2023/03/03 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

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.

Related

• Decentralized Adversarial Training over Graphs
• Stability and Generalization of Adversarial Diffusion Training
• On the Escaping Efficiency of Distributed Adversarial Training Algorithms
• Adversarial Training for Gradient Descent: Analysis Through its Continuous-time Approximation
• Distributed Statistical Min-Max Learning in the Presence of Byzantine Agents