arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2402.06388stat.MLcs.AIcs.DScs.LGcs.NAmath.NA
收藏

多臂老虎机中L2正则化策略梯度算法的收敛性

Convergence of a L2 regularized Policy Gradient Algorithm for the Multi Armed Bandit

``Octav Mayer'' Institute of Mathematics of the Romanian Academy Bd. Carol I 8, Ia s i 700505, Romania CEREMADE, Universit\'e Paris Dauphine - PSL, CNRS, Paris, France;anonymous institute

Stefana Anita, Gabriel Turinici

AI总结:

本文研究了多臂老虎机中结合L2正则化和softmax参数化的策略梯度算法的收敛性,证明了其在适当假设下的收敛性,并通过实验展示了其在初始猜测远离解时的优势。

摘要:

尽管多臂老虎机(MAB)和策略梯度方法是强化学习中最常用的框架之一,但用于MAB的策略梯度算法的理论性质并未得到足够关注。本文研究了在存在L2正则化项并与'softmax'参数化联合使用的情况下,此类过程的收敛性。我们在适当的技术假设下证明了收敛性,并通过数值测试验证了该过程,包括超出理论设定的情况。测试表明,时间依赖的正则化过程在初始猜测远离解时能优于传统方法。

英文摘要:

Although Multi Armed Bandit (MAB) on one hand and the policy gradient approach on the other hand are among the most used frameworks of Reinforcement Learning, the theoretical properties of the policy gradient algorithm used for MAB have not been given enough attention. We investigate in this work the convergence of such a procedure for the situation when a $L2$ regularization term is present jointly with the 'softmax' parametrization. We prove convergence under appropriate technical hypotheses and test numerically the procedure including situations beyond the theoretical setting. The tests show that a time dependent regularized procedure can improve over the canonical approach especially when the initial guess is far from the solution.

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。

↑