arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2011.12895cs.LGcs.AIcs.MA
收藏

TLeague:一个基于竞争性自我游戏的分布式多智能体强化学习框架

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

Tencent Robotics X(腾讯机器人X实验室);Tsinghua University(清华大学)

Peng Sun, Jiechao Xiong, Lei Han, Xinghai Sun, Shuxing Li, Jiawei Xu, Meng Fang, Zhengyou Zhang

摘要:

基于竞争自玩(CSP)的多智能体强化学习(MARL)近年来取得了惊人的突破。强大的AI在多个基准测试中获得,包括Dota 2,王者荣耀,雷神之锤III,星际争霸II,仅举几例。尽管取得了成功,但MARL训练非常需要数据,通常需要在训练期间从环境中看到数十亿(如果不是数万亿)帧,以便学习高性能代理。这给研究人员或工程师带来了不小的困难,并阻止了MARL在更广泛的现实问题中的应用。为了解决这个问题,在本文中,我们描述了一个框架,称为TLeague,旨在大规模的训练和实现几个主流的CSP-MARL算法。培训可以部署在单个机器或混合机器(CPU和GPU)集群中,其中以云原生方式支持标准Kubernetes。TLeague在执行分布式训练时实现了高吞吐量和合理的扩展。由于采用了模块化设计,该算法也易于扩展,用于解决其他多智能体问题或实现和验证MARL算法。我们在星际争霸II,ViZDoom和Pommerman上进行了实验,以显示TLeague的效率和有效性。该代码是开源的,可在https://github.com/tencent-ailab/tleague_projpage上获得

英文摘要:

Competitive Self-Play (CSP) based Multi-Agent Reinforcement Learning (MARL) has shown phenomenal breakthroughs recently. Strong AIs are achieved for several benchmarks, including Dota 2, Glory of Kings, Quake III, StarCraft II, to name a few. Despite the success, the MARL training is extremely data thirsty, requiring typically billions of (if not trillions of) frames be seen from the environment during training in order for learning a high performance agent. This poses non-trivial difficulties for researchers or engineers and prevents the application of MARL to a broader range of real-world problems. To address this issue, in this manuscript we describe a framework, referred to as TLeague, that aims at large-scale training and implements several main-stream CSP-MARL algorithms. The training can be deployed in either a single machine or a cluster of hybrid machines (CPUs and GPUs), where the standard Kubernetes is supported in a cloud native manner. TLeague achieves a high throughput and a reasonable scale-up when performing distributed training. Thanks to the modular design, it is also easy to extend for solving other multi-agent problems or implementing and verifying MARL algorithms. We present experiments over StarCraft II, ViZDoom and Pommerman to show the efficiency and effectiveness of TLeague. The code is open-sourced and available at https://github.com/tencent-ailab/tleague_projpage

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑