---
title: hands-on-modern-rl-experiment04-board-selfplay
canonical_url: "https://www.modelscope.cn/studios/walkinglab/hands-on-modern-rl-experiment04-board-selfplay"
md_url: "https://www.modelscope.cn/studios/walkinglab/hands-on-modern-rl-experiment04-board-selfplay.md"
repository: walkinglab/hands-on-modern-rl-experiment04-board-selfplay
chinese_name: "Experiment 04 · Board Games & Self-Play"
last_updated: 2026-08-20
sdk_type: gradio
sdk_version: 6.17.3
downloads: 0
stars: 0
---

# hands-on-modern-rl-experiment04-board-selfplay

> hands-on-modern-rl-experiment04-board-selfplay - walkinglab 在 ModelScope 创建的在线 Demo。Hands-On Modern RL 配套实验：CFR+、表格自博弈与完整棋局回放。

walkinglab/hands-on-modern-rl-experiment04-board-selfplay 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 04 · Board Games & Self-Play」。

- **Repository**: walkinglab/hands-on-modern-rl-experiment04-board-selfplay
- **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-experiment04-board-selfplay

---

# WalkingLab × Hands-On Modern RL · Board Games & Self-Play Lab

**WalkingLab** 与开源课程 **Hands-On Modern RL（《动手学现代强化学习》）** 的棋盘游戏实验。在 ModelScope CPU 容器中运行 OpenSpiel 的 CFR+ 与表格自博弈训练。页面持续显示均衡差距或对随机策略胜率，训练结束后生成完整对局回放。

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 中运行棋盘游戏实验](https://modelscope.cn/notebook/share/github/walkinglabs/hands-on-modern-rl/blob/main/code/online-experiments/hands-on-modern-rl-experiment04-board-selfplay.ipynb)。Notebook 与当前创空间复用同一份 CFR+ 与表格自博弈运行时，并显示训练日志、策略指标和结果回放。
