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 2102.01585cs.MAcs.AIcs.LG
收藏

基于熵正则化深度强化学习的平均场博弈近似求解

Approximately Solving Mean Field Games via Entropy-Regularized Deep Reinforcement Learning

Technische Universität Darmstadt(达姆施塔特工业大学)

Kai Cui, Heinz Koeppl

摘要:

最近的平均场博弈(MFG)的形式主义,否则难以计算的近似纳什均衡在许多代理设置。在本文中,我们考虑离散时间有限MFG有限时域目标。我们发现,所有的离散时间有限MFG与非常数不动点算子未能压缩通常假设在现有的MFG文献,除非通过不动点迭代收敛。相反,我们将熵正则化和玻尔兹曼政策到不动点迭代。结果,我们获得了可证明的收敛到现有方法失败的近似不动点,并达到了近似纳什均衡的原始目标。所有提出的方法进行评估,就其可利用性,在两个有指导意义的例子与易于处理的精确解和高维问题的精确方法变得棘手。在高维场景中,我们应用已建立的深度强化学习方法,并根据经验将虚拟游戏与我们的近似值结合起来。

英文摘要:

The recent mean field game (MFG) formalism facilitates otherwise intractable computation of approximate Nash equilibria in many-agent settings. In this paper, we consider discrete-time finite MFGs subject to finite-horizon objectives. We show that all discrete-time finite MFGs with non-constant fixed point operators fail to be contractive as typically assumed in existing MFG literature, barring convergence via fixed point iteration. Instead, we incorporate entropy-regularization and Boltzmann policies into the fixed point iteration. As a result, we obtain provable convergence to approximate fixed points where existing methods fail, and reach the original goal of approximate Nash equilibria. All proposed methods are evaluated with respect to their exploitability, on both instructive examples with tractable exact solutions and high-dimensional problems where exact methods become intractable. In high-dimensional scenarios, we apply established deep reinforcement learning methods and empirically combine fictitious play with our approximations.

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。

↑