BriefGPT.xyz
Mar, 2023
基于图的去中心化对抗训练
Decentralized Adversarial Training over Graphs
HTML
PDF
Ying Cao, Elsa Rizk, Stefan Vlaski, Ali H. Sayed
TL;DR
研究在图形上进行分散敌对训练以提高多智能体系统的鲁棒性。通过使用传播学习的 min-max 形式，我们开发了一种分散的敌对训练框架，在凸和非凸环境中分析了所提出方案的收敛性，并说明了对抗攻击的增强鲁棒性。
Abstract
The vulnerability of
machine learning models
to
adversarial attacks
has been attracting considerable attention in recent years. Most existing studies focus on the behavior of stand-alone single-agent learners. In
→