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 2110.09712cs.LG
收藏

用现实主义的演员-评论家来平衡价值低估和高估

Balancing Value Underestimation and Overestimation with Realistic Actor-Critic

Harbin Engineering University(哈尔滨工程大学)

Sicen Li, Qinyun Tang, Yiming Pang, Xinmeng Ma, Gang Wang

摘要:

无模型深度强化学习(RL)已成功应用于具有挑战性的连续控制领域。然而,差的样本效率阻碍了这些方法在现实世界领域中的广泛应用。本文介绍了一种新的无模型算法,现实演员-评论家(RAC),它可以与任何非政策RL算法,以提高样本效率。RAC采用通用价值函数逼近器(UVFA)来同时学习具有相同神经网络的策略族,每个策略族在低估和高估之间具有不同的权衡。为了学习此类政策,我们引入了不确定性惩罚Q学习,它利用多个批评者集合的不确定性来建立Q函数的各种置信区间。我们在MuJoCo基准上评估RAC,与SAC相比,在最具挑战性的Humanoid环境中实现了10倍样本效率和25%的性能改进。

英文摘要:

Model-free deep reinforcement learning (RL) has been successfully applied to challenging continuous control domains. However, poor sample efficiency prevents these methods from being widely used in real-world domains. This paper introduces a novel model-free algorithm, Realistic Actor-Critic(RAC), which can be incorporated with any off-policy RL algorithms to improve sample efficiency. RAC employs Universal Value Function Approximators (UVFA) to simultaneously learn a policy family with the same neural network, each with different trade-offs between underestimation and overestimation. To learn such policies, we introduce uncertainty punished Q-learning, which uses uncertainty from the ensembling of multiple critics to build various confidence-bounds of Q-function. We evaluate RAC on the MuJoCo benchmark, achieving 10x sample efficiency and 25\% performance improvement on the most challenging Humanoid environment compared to SAC.

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。

↑