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 1906.09734cs.LGstat.ML
收藏

第一人称射击游戏中经验的最佳利用

Optimal Use of Experience in First Person Shooter Environments

College of Engineering and Computer Science(工程与计算机科学学院);Australian National University(澳大利亚国立大学)

Matthew Aitchison

摘要:

尽管强化学习最近取得了长足的进步,但它仍然存在一个局限性,即它需要与环境进行极高次数的交互。在本文中,我们探讨了在深度Q-学习算法中重用来自经验重放缓冲区的经验的有效性。我们测试了在VizDoom环境中每个环境步骤多次应用学习更新步骤的有效性,首先,这需要改变学习率,其次,它不会提高代理的性能。此外,我们表明,更新频率较低是有效的比例为4:1,之后的性能显着下降。这些结果定量地证实了每第四个环境步骤执行学习更新的广泛实践。

英文摘要:

Although reinforcement learning has made great strides recently, a continuing limitation is that it requires an extremely high number of interactions with the environment. In this paper, we explore the effectiveness of reusing experience from the experience replay buffer in the Deep Q-Learning algorithm. We test the effectiveness of applying learning update steps multiple times per environmental step in the VizDoom environment and show first, this requires a change in the learning rate, and second that it does not improve the performance of the agent. Furthermore, we show that updating less frequently is effective up to a ratio of 4:1, after which performance degrades significantly. These results quantitatively confirm the widespread practice of performing learning updates every 4th environmental step.

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。

↑