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 2303.09629cs.LGcs.AI
收藏

周期性MDP模型的在线强化学习

Online Reinforcement Learning in Periodic MDP

IIT Delhi(印度理工学院德里分校)

Ayush Aniket, Arpan Chattopadhyay

摘要:

我们研究周期性马尔可夫决策过程(MDP)中的学习,这是一种特殊类型的非平稳MDP,其中状态转移概率和奖励函数在平均奖励最大化设置下周期性变化。通过在状态空间中增加周期指标,将问题转化为一个平稳的MDP问题,并提出了一个周期置信上限强化学习-2(PUCRL 2)算法.结果表明,PUCRL 2的遗憾度随周期$N$线性变化,$\mathcal{O}(\sqrt{Tlog T})$随视界长度$T$线性变化.利用增广MDP的转移矩阵的稀疏性的信息,我们提出了另一种算法PUCRLB的PUCRL 2的增强,无论是在遗憾($O(\sqrt{N})$依赖于周期)和经验性能。最后,我们提出了另外两个算法U-PUCRL 2和U-PUCRLB扩展的不确定性的环境中,周期是未知的,但一组候选周期是已知的。数值结果表明了所有算法的有效性。

英文摘要:

We study learning in periodic Markov Decision Process (MDP), a special type of non-stationary MDP where both the state transition probabilities and reward functions vary periodically, under the average reward maximization setting. We formulate the problem as a stationary MDP by augmenting the state space with the period index, and propose a periodic upper confidence bound reinforcement learning-2 (PUCRL2) algorithm. We show that the regret of PUCRL2 varies linearly with the period $N$ and as $\mathcal{O}(\sqrt{Tlog T})$ with the horizon length $T$. Utilizing the information about the sparsity of transition matrix of augmented MDP, we propose another algorithm PUCRLB which enhances upon PUCRL2, both in terms of regret ($O(\sqrt{N})$ dependency on period) and empirical performance. Finally, we propose two other algorithms U-PUCRL2 and U-PUCRLB for extended uncertainty in the environment in which the period is unknown but a set of candidate periods are known. Numerical results demonstrate the efficacy of all the algorithms.

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。

↑