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 2307.07014cs.LGcs.AIstat.ML
收藏

利用因子化动作空间进行离策略评估

Leveraging Factored Action Spaces for Off-Policy Evaluation

Imperial College London(伦敦帝国学院);University of Michigan(密歇根大学)

Aaman Rebello, Shengpu Tang, Jenna Wiens, Sonali Parbhoo

AI总结:

本文提出基于因子化动作空间的分解式重要性采样 OPE 估计量,在保持零偏差的同时降低方差,并通过模拟验证其理论优势。

摘要:

离策略评估(OPE)旨在根据从已执行动作序列中收集的数据,估计遵循反事实动作序列所能带来的收益。然而,在涉及大型组合式动作空间的问题中,现有 OPE 估计量往往表现出高偏差和高方差。我们研究如何利用因子化动作空间来缓解这一问题,即将每个动作表示为来自较小动作空间的独立子动作的组合。该方法有助于更细粒度地分析各动作在效果上的差异。在本工作中,我们基于因子化动作空间提出了一族新的“分解式”重要性采样(IS)估计量。在对底层问题结构作出某些假设的条件下,我们证明分解式 IS 估计量相比其原始未分解版本具有更低方差,同时保持零偏差特性。通过模拟,我们从经验上验证了理论结果,并考察了各类假设的有效性。在提供一种能够为给定问题推导动作空间因子分解的技术后,本研究表明,利用这种内在问题结构可以“免费”改进 OPE。

英文摘要:

Off-policy evaluation (OPE) aims to estimate the benefit of following a counterfactual sequence of actions, given data collected from executed sequences. However, existing OPE estimators often exhibit high bias and high variance in problems involving large, combinatorial action spaces. We investigate how to mitigate this issue using factored action spaces i.e. expressing each action as a combination of independent sub-actions from smaller action spaces. This approach facilitates a finer-grained analysis of how actions differ in their effects. In this work, we propose a new family of "decomposed" importance sampling (IS) estimators based on factored action spaces. Given certain assumptions on the underlying problem structure, we prove that the decomposed IS estimators have less variance than their original non-decomposed versions, while preserving the property of zero bias. Through simulations, we empirically verify our theoretical results, probing the validity of various assumptions. Provided with a technique that can derive the action space factorisation for a given problem, our work shows that OPE can be improved "for free" by utilising this inherent problem structure.

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。

↑