Collaborative Computing in Multi UAV MEC Network OptimizationMulti-UAV collaborative path planing.
为什么不能用增强学习的方法?因为以下三点风险:GIoT的设备会动态地加入或者离开网络。恶劣的用户会上传低质量甚至有害的数据。甚至会发动逆模型攻击:即通过训练好的模型来提取训练数据。
MADIFF: Offline Multi-agent Learning with Diffusion ModelsMADIFF is the ffrst diffusion-based multi-agent learning framework, which behaves as both a decentralized policy and a centralized controller.
下面这句话说明了为什么diffusion model用于多智能体非监督式学习的创新点。Despite its effectiveness in single-agent learning, applying the g ...