arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2008.08932cs.LGcs.AI
收藏

SuperSuit:用于强化学习环境的简单微包装器

SuperSuit: Simple Microwrappers for Reinforcement Learning Environments

University of Maryland, College Park(马里兰大学帕克分校)

J. K. Terry, Benjamin Black, Ananth Hari

摘要:

在强化学习中,包装器普遍用于转换模型和环境之间传递的信息。尽管它们无处不在,但没有一个库能够合理地实现所有流行的预处理方法。这会导致不必要的bug、代码效率低下和浪费开发人员的时间。因此,我们引入了SuperSuit,一个包含所有流行包装器的Python库,以及可以轻松将lambda函数应用于观察/动作/奖励的包装器。它与标准的Gym环境规范以及多代理环境的PettingZoo规范兼容。这个库可以在https://github.com/PettingZoo-Team/SuperSuit上找到,并且可以通过pip安装。

英文摘要:

In reinforcement learning, wrappers are universally used to transform the information that passes between a model and an environment. Despite their ubiquity, no library exists with reasonable implementations of all popular preprocessing methods. This leads to unnecessary bugs, code inefficiencies, and wasted developer time. Accordingly we introduce SuperSuit, a Python library that includes all popular wrappers, and wrappers that can easily apply lambda functions to the observations/actions/reward. It's compatible with the standard Gym environment specification, as well as the PettingZoo specification for multi-agent environments. The library is available at https://github.com/PettingZoo-Team/SuperSuit,and can be installed via pip.

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。

↑