BriefGPT.xyz
Apr, 2022
随机博弈中的马尔科夫均衡复杂性
The Complexity of Markov Equilibrium in Stochastic Games
HTML
PDF
Constantinos Daskalakis, Noah Golowich, Kaiqing Zhang
TL;DR
本文研究了Markov粗粒度关联均衡问题的计算复杂性及其在多智能体强化学习中的应用,发现当多智能体交互为回合制、折扣因子和粗略程度为常数时,计算近似的Markov粗粒度关联均衡策略属于NP难问题,但是提供了在多智能体中非稳定Markov CCE策略的学习解决方案。
Abstract
We show that computing approximate stationary
markov coarse correlated equilibria
(CCE) in general-sum
stochastic games
is computationally intractable, even when there are two players, the game is turn-based, the
→