arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2502.04477math.OCcs.DS
收藏

通过锚定法实现马尔可夫决策过程的近最优样本复杂度

Near-Optimal Sample Complexity for MDPs via Anchoring

Jongmin Lee, Mario Bravo, Roberto Cominetti

AI总结:

本研究提出结合递归采样与Halpern锚定迭代的无模型算法,在弱通信平均奖励MDPs中实现了近最优的样本与时间复杂度,无需先验知识且保证有限时间终止,还可拓展至折扣MDPs场景。

摘要:

我们研究了一种在弱通信情形下,为平均奖励马尔可夫决策过程(MDPs)计算ε-最优策略的新型无模型算法。在给定生成模型的条件下,我们的方法将递归采样技术与Halpern锚定迭代相结合,计算ε-最优策略的样本复杂度和时间复杂度均为$\tilde{O}(|\fancyscript{S}||\fancyscript{A}|\fancyscript{h}^*_{sp}^2/\fancyscript{\u0025}^2}$,该结果在高概率和期望意义下均成立。据我们所知,这是无模型算法中目前最优的复杂度结果,与已知下界的差距仅为$\fancyscript{h}^*_{sp}$因子。尽管该复杂度界涉及未知偏差向量的跨度半范数$\fancyscript{h}^*_{sp}$,但算法无需任何先验知识,且采用了一种以概率1保证有限时间终止的停止规则。我们还分析了这些技术如何适配折扣MDPs的场景。

英文摘要:

We study a new model-free algorithm to compute $\varepsilon$-optimal policies for average reward Markov decision processes, in the weakly communicating case. Given a generative model, our procedure combines a recursive sampling technique with Halpern's anchored iteration, and computes an $\varepsilon$-optimal policy with sample and time complexity $\widetilde{O}(|\mathcal{S}||\mathcal{A}|\|h^*\|_{\text{sp}}^{2}/\varepsilon^{2})$ both in high probability and in expectation. To our knowledge, this is the best complexity among model-free algorithms, matching the known lower bound up to a factor $\|h^*\|_{\text{sp}}$. Although the complexity bound involves the span seminorm $\|h^*\|_{\text{sp}}$ of the unknown bias vector, the algorithm requires no prior knowledge and implements a stopping rule which guarantees with probability 1 that the procedure terminates in finite time. We also analyze how these techniques can be adapted for discounted MDPs.

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。

↑