arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2203.11973cs.LGmath.OCstat.ML
收藏

平均场博弈的可扩展深度强化学习算法

Scalable Deep Reinforcement Learning Algorithms for Mean Field Games

Google Research(谷歌研究院);Univ. Lille(里尔大学);CNRS(法国国家科学研究中心);Inria(法国国家信息与自动化研究所);Centrale Lille(里尔中央理工学院);DeepMind(深度思维);UC Berkeley(加州大学伯克利分校);Singapore University of Technology and Design(新加坡科技设计大学)

Mathieu Laurière, Sarah Perrin, Sertan Girgin, Paul Muller, Ayush Jain, Theophile Cabannes, Georgios Piliouras, Julien Pérolat, Romuald Élie, Olivier Pietquin, Matthieu Geist

摘要:

平均场游戏(MFG)已被引入到有效地近似游戏非常大的人口的战略代理。最近,在MFG的学习均衡的问题已经获得了动力,特别是使用无模型强化学习(RL)方法。使用RL进一步扩大规模的一个限制因素是求解MFG的现有算法需要混合近似量,例如策略或$q$值。这在具有良好泛化特性的非线性函数逼近(例如神经网络)的情况下远非微不足道。我们提出了两种方法来解决这个问题。第一个学习的混合策略,从蒸馏的历史数据到一个神经网络,并应用于虚构游戏算法。第二种是基于正则化的在线混合方法,不需要记住历史数据或以前的估计。它用于扩展Online Mirror Descent。我们在数值上证明了这些方法可以有效地使用Deep RL算法来解决各种MFG。此外,我们表明,这些方法优于SotA基线的文献。

英文摘要:

Mean Field Games (MFGs) have been introduced to efficiently approximate games with very large populations of strategic agents. Recently, the question of learning equilibria in MFGs has gained momentum, particularly using model-free reinforcement learning (RL) methods. One limiting factor to further scale up using RL is that existing algorithms to solve MFGs require the mixing of approximated quantities such as strategies or $q$-values. This is far from being trivial in the case of non-linear function approximation that enjoy good generalization properties, e.g. neural networks. We propose two methods to address this shortcoming. The first one learns a mixed strategy from distillation of historical data into a neural network and is applied to the Fictitious Play algorithm. The second one is an online mixing method based on regularization that does not require memorizing historical data or previous estimates. It is used to extend Online Mirror Descent. We demonstrate numerically that these methods efficiently enable the use of Deep RL algorithms to solve various MFGs. In addition, we show that these methods outperform SotA baselines from the literature.

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。

↑