arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2006.08906cs.LGcs.AIstat.ML
收藏

多样本有效时间差分学习的元学习资格迹

META-Learning Eligibility Traces for More Sample Efficient Temporal Difference Learning

McGill University(麦吉尔大学)

Mingde Zhao

摘要:

时间差分(TD)学习是一种标准的、非常成功的强化学习方法,是学习给定策略值的算法以及学习如何改进策略的算法的核心。具有资格跟踪的TD学习提供了一种进行时间信用分配的方式,即,决定奖励的哪一部分应该被分配给在不同的先前时间发生的前驱状态,由参数$λ$控制。但是,调整此参数可能很耗时,不调整它可能会导致学习效率低下。为了提高TD学习的样本效率,我们提出了一种元学习方法来调整资格跟踪参数,在一个状态依赖的方式。该自适应是在辅助学习器的帮助下实现的,辅助学习器在线学习关于更新目标的分布信息,每步产生与通常的值学习器大致相同的计算复杂性。我们的方法可以用于在政策和离策略学习。我们证明,在一些假设下,该方法提高了更新目标的整体质量,通过最小化整体目标误差。该方法可以被看作是一个插件,也可以用来辅助预测与功能近似的元学习功能(观察)为基础的$λ$在线,甚至在控制的情况下,以协助政策的改进。我们的实证评估表明显着的性能改进,以及改进的鲁棒性所提出的算法的学习率变化。

英文摘要:

Temporal-Difference (TD) learning is a standard and very successful reinforcement learning approach, at the core of both algorithms that learn the value of a given policy, as well as algorithms which learn how to improve policies. TD-learning with eligibility traces provides a way to do temporal credit assignment, i.e. decide which portion of a reward should be assigned to predecessor states that occurred at different previous times, controlled by a parameter $λ$. However, tuning this parameter can be time-consuming, and not tuning it can lead to inefficient learning. To improve the sample efficiency of TD-learning, we propose a meta-learning method for adjusting the eligibility trace parameter, in a state-dependent manner. The adaptation is achieved with the help of auxiliary learners that learn distributional information about the update targets online, incurring roughly the same computational complexity per step as the usual value learner. Our approach can be used both in on-policy and off-policy learning. We prove that, under some assumptions, the proposed method improves the overall quality of the update targets, by minimizing the overall target error. This method can be viewed as a plugin which can also be used to assist prediction with function approximation by meta-learning feature (observation)-based $λ$ online, or even in the control case to assist policy improvement. Our empirical evaluation demonstrates significant performance improvements, as well as improved robustness of the proposed algorithm to learning rate variation.

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。

↑