**摘要**
协作多目标多智能体强化学习( MOMARL )在多个潜在冲突的目标下对团队决策进行建模。在这种情况下,冲突不仅发生在目标之间,也发生在具有不同观察结果、角色和贡献的代理之间。我们提出了偏好协调多智能体策略优化( PCMA ) ,它学习协调智能体
👤 作者: Pengxin Wang, Lihao Guo, Yi Xie, Bo Liu, Siyang Cao, Jingdi Chen
---
🔗 **[Learning Coordinated Preference for Multi-Objective Multi-Agent Reinforcement Learning](https://arxiv.org/abs/2606.14693v1)**
> Learning Coordinated Preference for Multi-Objective Multi-Agent Reinforcement Learning
🏷️ 来源: ArXiv cs.AI
⏱️ 2026-06-15 14:00
news
Learning Coordinated Preference for Multi-Objective Multi-Agent Reinforcement Learning
💬 评论
讨论话题: 你愿意花钱雇一个AI Agent干活吗?如果可以,你愿意付多少钱?你觉得什么样的AI服务你会心甘情愿付费?
Loading replies...
加载评论中...