用强化学习玩《Generals.io》，36小时登顶全球前万分之三。
Artificial Generals Intelligence: Mastering Generals.io with Reinforcement Learning
- 基于真实游戏构建可复现的多人实时战略环境，支持高速训练。
- 单卡H100训练36小时，击败99.997%人类玩家，达顶尖水平。
- 适合研究多智能体强化学习，代码开源，易上手。
我们基于广受欢迎的实时战略游戏 Generals.io 构建了一个全新的游戏环境，该环境兼容 Gymnasium 与 PettingZoo，可在普通硬件上每秒运行数千帧。我们提出一个基准智能体，通过监督预训练与自对弈训练，在单张 H100 GPU 上仅用36小时便达到 1v1 人类排行榜前 0.003% 的水平。为加速学习，引入基于潜在函数的奖励塑形与记忆特征。我们的模块化 RTS 基准与高性能基线智能体，为多智能体强化学习研究提供了既易访问又具挑战性的平台。完整代码、示例与教程已公开于 https://github.com/strakam/generals-bots。
原文摘要 · Abstract (English)
We introduce a real-time strategy game environment based on Generals.io, a game with thousands of weekly active players. Our environment is fully compatible with Gymnasium and PettingZoo and is capable of running thousands of frames per second on commodity hardware. We also present a reference agent, trained with supervised pre-training and self-play, which reached the top 0.003% of the 1v1 human leaderboard after only 36 hours on a single H100 GPU. To accelerate learning, we incorporate potential-based reward shaping and memory features. Our contributions of a modular RTS benchmark and a competitive baseline agent provide an accessible yet challenging platform for advancing multi-agent reinforcement learning research. The documented code, together with examples and tutorials, is available at https://github.com/strakam/generals-bots.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。