---
title: hands-on-modern-rl-experiment05-multiagent-games
canonical_url: "https://www.modelscope.cn/studios/walkinglab/hands-on-modern-rl-experiment05-multiagent-games"
md_url: "https://www.modelscope.cn/studios/walkinglab/hands-on-modern-rl-experiment05-multiagent-games.md"
repository: walkinglab/hands-on-modern-rl-experiment05-multiagent-games
chinese_name: "Experiment 05 · Multi-Agent Games"
last_updated: 2026-08-20
sdk_type: gradio
sdk_version: 6.17.3
downloads: 0
stars: 0
---

# hands-on-modern-rl-experiment05-multiagent-games

> hands-on-modern-rl-experiment05-multiagent-games - walkinglab 在 ModelScope 创建的在线 Demo。Hands-On Modern RL 配套实验：参数共享 PPO、合作与竞争游戏、同步回放。

walkinglab/hands-on-modern-rl-experiment05-multiagent-games 是 ModelScope 魔搭社区上的在线可交互 Demo（创空间），基于 gradio 6.17.3 构建，中文名为「Experiment 05 · Multi-Agent Games」。

- **Repository**: walkinglab/hands-on-modern-rl-experiment05-multiagent-games
- **SDK**: gradio
- **SDK version**: 6.17.3
- **Downloads**: 0
- **Stars**: 0
- **Last updated**: 2026-08-20

Source: https://www.modelscope.cn/studios/walkinglab/hands-on-modern-rl-experiment05-multiagent-games

---

# WalkingLab × Hands-On Modern RL · Multi-Agent CPU Game Arena

**WalkingLab** 与开源课程 **Hands-On Modern RL（《动手学现代强化学习》）** 的多智能体实验。在 ModelScope CPU 容器中运行 PettingZoo 合作与竞争游戏，使用参数共享 PPO 训练多个智能体，并生成同步策略回放。

课程项目：<https://github.com/walkinglabs/hands-on-modern-rl>

WalkingLab：<https://modelscope.cn/organization/walkinglab>

## 配套实验 Notebook

[直接在 ModelScope Notebook 中运行多智能体实验](https://modelscope.cn/notebook/share/github/walkinglabs/hands-on-modern-rl/blob/main/code/online-experiments/hands-on-modern-rl-experiment05-multiagent-games.ipynb)。Notebook 与当前创空间复用同一份 PettingZoo/PPO 训练运行时，并显示完整日志、团队奖励曲线和同步策略回放。