arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 1909.12397cs.LGcs.AIstat.ML
收藏

CAQL:持续行动Q学习

CAQL: Continuous Action Q-Learning

Google Research(谷歌研究院)

Moonkyung Ryu, Yinlam Chow, Ross Anderson, Christian Tjandraatmadja, Craig Boutilier

摘要:

基于值的强化学习(RL)方法(如Q学习)在许多领域都取得了成功。然而,将Q学习应用于连续动作RL问题的一个挑战是最佳Bellman备份所需的连续动作最大化(max-Q)。在这项工作中,我们开发了CAQL,这是一种用于连续动作Q学习的算法,可以使用几个即插即用的优化器来解决最大Q问题。利用深度神经网络的最新优化结果,我们证明了可以使用混合整数规划(MIP)来最佳地解决max-Q。当Q函数表示具有足够的功率时,基于MIP的优化产生更好的策略并且比近似方法(例如,梯度上升、交叉熵搜索)。我们进一步开发了几种技术来加速CAQL中的推理,尽管它们具有近似的性质,但性能良好。我们比较CAQL与最先进的RL算法的基准连续控制问题,有不同程度的行动约束,并表明CAQL优于基于政策的方法在严重约束的环境中,往往显着。

英文摘要:

Value-based reinforcement learning (RL) methods like Q-learning have shown success in a variety of domains. One challenge in applying Q-learning to continuous-action RL problems, however, is the continuous action maximization (max-Q) required for optimal Bellman backup. In this work, we develop CAQL, a (class of) algorithm(s) for continuous-action Q-learning that can use several plug-and-play optimizers for the max-Q problem. Leveraging recent optimization results for deep neural networks, we show that max-Q can be solved optimally using mixed-integer programming (MIP). When the Q-function representation has sufficient power, MIP-based optimization gives rise to better policies and is more robust than approximate methods (e.g., gradient ascent, cross-entropy search). We further develop several techniques to accelerate inference in CAQL, which despite their approximate nature, perform well. We compare CAQL with state-of-the-art RL algorithms on benchmark continuous-control problems that have different degrees of action constraints and show that CAQL outperforms policy-based methods in heavily constrained environments, often dramatically.

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。

↑