当前位置: 代码迷 >> 综合 >> TLeague: A Framework for Competitive Self-Play based Distributed Multi-Agent Reinforcement Learning
  详细解决方案

TLeague: A Framework for Competitive Self-Play based Distributed Multi-Agent Reinforcement Learning

热度:42   发布时间:2023-12-12 08:47:28.0

文章目录

  • 前言
  • Architecture
    • Mathematical Settings
    • Design
      • Actor
      • Learner
      • InfServer
      • ModelPool
      • LeagueMgr
    • System-Level Design
    • Large-scale Run and Kubernetes
    • Code Structure
      • Arena
      • TPolicies
      • TLeague
    • Extension
      • Adding New Env
      • Adding
  相关解决方案