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 2410.02605cs.LGcs.AI
收藏

基于累积前景理论的强化学习策略梯度

Policy Gradients for Cumulative Prospect Theory in Reinforcement Learning

department of computer science at ETH Zürich(苏黎世联邦理工学院计算机科学系);Singapore University of Technology and Design(新加坡科技与设计大学)

Olivier Lepel, Anas Barakat

AI总结:

本文提出了一种基于累积前景理论的强化学习策略梯度算法,通过蒙特卡洛梯度估计器实现,并证明了算法在CPT目标上的渐近收敛性。

摘要:

我们推导了有限时间 horizon 强化学习(RL)中累积前景理论(CPT)目标的策略梯度定理,推广了标准策略梯度定理,并涵盖了基于扭曲的风险目标作为特殊情况。受行为经济学启发,CPT结合了围绕参考点的非对称效用转换与概率扭曲。基于我们的定理,我们设计了一个一阶策略梯度算法用于CPT-RL,使用基于顺序统计的蒙特卡洛梯度估计器。我们建立了估计器的统计保证,并证明了所得到的算法在一般非凸CPT目标的首次阶 stationary 点上渐近收敛。模拟展示了CPT引起的定性行为,并将我们的首次阶方法与现有零阶方法进行了比较。

英文摘要:

We derive a policy gradient theorem for Cumulative Prospect Theory (CPT) objectives in finite-horizon Reinforcement Learning (RL), generalizing the standard policy gradient theorem and encompassing distortion-based risk objectives as special cases. Motivated by behavioral economics, CPT combines an asymmetric utility transformation around a reference point with probability distortion. Building on our theorem, we design a first-order policy gradient algorithm for CPT-RL using a Monte Carlo gradient estimator based on order statistics. We establish statistical guarantees for the estimator and prove asymptotic convergence of the resulting algorithm to first-order stationary points of the (generally nonconvex) CPT objective. We complement our asymptotic analysis with a non-asymptotic total sample complexity analysis to reach an approximate first-order stationary policy. Simulations illustrate qualitative behaviors induced by CPT and compare our first-order approach to existing zeroth-order methods.

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。

↑