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 1910.09986cs.AI
收藏

通过将高级知识嵌入深度强化学习更快更安全地进行训练

Faster and Safer Training by Embedding High-Level Knowledge into Deep Reinforcement Learning

Shenzhen University(深圳大学);Shanghai Research Center for Brain Science and Brain-Inspired Intelligence(上海脑科学与类脑研究中心);Zhangjiang Laboratory(张江实验室);Dorabot Inc.(达闼机器人有限公司);Hong Kong University of Science and Technology(香港科技大学)

Haodi Zhang, Zihang Gao, Yi Zhou, Hao Zhang, Kaishun Wu, Fangzhen Lin

摘要:

深度强化学习已成功应用于许多动态决策领域,特别是那些具有非常大的状态空间的领域。然而,众所周知,深度强化学习可能非常缓慢且资源密集。由此产生的系统往往是脆弱的,难以解释。在本文中,我们试图通过提出一个规则插入学习(RIL)框架,将高级规则嵌入到深度强化学习中来解决其中的一些问题。有了一些好的规则,这个框架不仅可以加速学习过程,还可以避免灾难性的探索,从而使系统即使在训练的早期阶段也相对稳定。此外,由于规则是高级的,易于解释,因此可以轻松地维护,更新和与其他类似任务共享。

英文摘要:

Deep reinforcement learning has been successfully used in many dynamic decision making domains, especially those with very large state spaces. However, it is also well-known that deep reinforcement learning can be very slow and resource intensive. The resulting system is often brittle and difficult to explain. In this paper, we attempt to address some of these problems by proposing a framework of Rule-interposing Learning (RIL) that embeds high level rules into the deep reinforcement learning. With some good rules, this framework not only can accelerate the learning process, but also keep it away from catastrophic explorations, thus making the system relatively stable even during the very early stage of training. Moreover, given the rules are high level and easy to interpret, they can be easily maintained, updated and shared with other similar tasks.

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。

↑