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 2202.02693cs.LGcs.AI
收藏

分布式强化学习的多样本目标值探索

Exploration with Multi-Sample Target Values for Distributional Reinforcement Learning

University of Oxford(牛津大学);University of British Columbia(英属哥伦比亚大学);Montréal Institute for Learning Algorithms(蒙特利尔学习算法研究所)

Michael Teng, Michiel van de Panne, Frank Wood

摘要:

分布式强化学习(RL)旨在学习一个价值网络,该网络预测给定状态的收益的完全分布,通常通过基于分位数的批评来建模。这种方法已经成功地集成到常见的RL方法中,用于连续控制,从而产生了分布式软演员-批评(DSAC)等算法。在本文中,我们介绍了多样本目标值(MTV)的分布RL,作为一个原则性的替代单样本目标值估计,在目前的实践中普遍采用。改进后的分布估计进一步有助于基于UCB的勘探。这两个想法结合在一起,产生了我们的分布式RL算法E2 DC(Extra Exploration with Distributional Critics)。我们在一系列连续控制任务上评估了我们的方法,并在困难的任务(如人形机器人控制)上展示了最先进的无模型性能。我们通过可视化和分析学习的分布及其在训练过程中的演变来进一步了解该方法。

英文摘要:

Distributional reinforcement learning (RL) aims to learn a value-network that predicts the full distribution of the returns for a given state, often modeled via a quantile-based critic. This approach has been successfully integrated into common RL methods for continuous control, giving rise to algorithms such as Distributional Soft Actor-Critic (DSAC). In this paper, we introduce multi-sample target values (MTV) for distributional RL, as a principled replacement for single-sample target value estimation, as commonly employed in current practice. The improved distributional estimates further lend themselves to UCB-based exploration. These two ideas are combined to yield our distributional RL algorithm, E2DC (Extra Exploration with Distributional Critics). We evaluate our approach on a range of continuous control tasks and demonstrate state-of-the-art model-free performance on difficult tasks such as Humanoid control. We provide further insight into the method via visualization and analysis of the learned distributions and their evolution during training.

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。

↑