arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2402.07082cs.LGcs.GTstat.ML
收藏

采用独立线性函数逼近的 Markov 博弈的精细样本复杂度

Refined Sample Complexity for Markov Games with Independent Linear Function Approximation

IIIS, Tsinghua University(清华大学交叉信息研究院);University of Washington(华盛顿大学)

Yan Dai, Qiwen Cui, Simon S. Du

AI总结:

本文改进 AVLPR 框架并引入动作相关奖励,为独立线性函数逼近下的 Markov 博弈提出首个兼具 $O(T^{-1/2})$ 最优速率且无 $A_{\max}$ 多项式依赖的多智能体 RL 算法。

摘要:

Markov Games(MG,Markov 博弈)是 Multi-Agent Reinforcement Learning(MARL,多智能体强化学习)中的重要模型。长期以来,人们一直认为“多智能体诅咒”(即算法性能随智能体数量呈指数级下降)不可避免,直到近期若干工作才改变这一看法(Daskalakis 等,2023;Cui 等,2023;Wang 等,2023)。尽管这些工作解决了多智能体诅咒,但当状态空间极大并采用(线性)函数逼近时,它们要么具有较慢的 $O(T^{-1/4})$ 收敛速率,要么带来对动作数量 $A_{\max}$ 的多项式依赖;而在单智能体情形中,即使损失函数可随时间任意变化,这种依赖也是可以避免的。本文首先改进 Wang 等(2023)的 AVLPR 框架,其核心洞见是设计依赖于数据的(即随机的)次优差距悲观估计,从而允许更广泛地选择可插入算法。当专门针对具有独立线性函数逼近的 MG 时,我们提出新颖的动作相关奖励,以覆盖偶发的极端估计误差。借助单智能体 RL 文献中的最先进技术,我们给出首个同时克服多智能体诅咒、达到最优 $O(T^{-1/2})$ 收敛速率并避免 $\text{poly}(A_{\max})$ 依赖的算法。

英文摘要:

Markov Games (MG) is an important model for Multi-Agent Reinforcement Learning (MARL). It was long believed that the "curse of multi-agents" (i.e., the algorithmic performance drops exponentially with the number of agents) is unavoidable until several recent works (Daskalakis et al., 2023; Cui et al., 2023; Wang et al., 2023). While these works resolved the curse of multi-agents, when the state spaces are prohibitively large and (linear) function approximations are deployed, they either had a slower convergence rate of $O(T^{-1/4})$ or brought a polynomial dependency on the number of actions $A_{\max}$ -- which is avoidable in single-agent cases even when the loss functions can arbitrarily vary with time. This paper first refines the AVLPR framework by Wang et al. (2023), with an insight of designing *data-dependent* (i.e., stochastic) pessimistic estimation of the sub-optimality gap, allowing a broader choice of plug-in algorithms. When specialized to MGs with independent linear function approximations, we propose novel *action-dependent bonuses* to cover occasionally extreme estimation errors. With the help of state-of-the-art techniques from the single-agent RL literature, we give the first algorithm that tackles the curse of multi-agents, attains the optimal $O(T^{-1/2})$ convergence rate, and avoids $\text{poly}(A_{\max})$ dependency simultaneously.

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。

↑