Ho-Bin Choi, Hyun-Kyo Lim, Gyu-Young Hwang, Youn-Hee Han,

Journal of Advanced Technology Research, Vol. 5, No. 2, pp. 19-23, Dec. 2020
10.11111/JATR.2020.5.2.019, Full Text:
Keywords: Reinforcement Learning, Reinforcement Learning Environment, Multi-agent
Abstract

Recently, AI technology has been applied in various places throughout society. Based on the continuous improvement of computing power, the advancement of AI technology will be accelerated and permeated more rapidly into human life. Among AI technologies, reinforcement learning is suitable for expressing and solving real-world problems. In fact, research on reinforcement learning that can solve real world problems is being actively conducted, and numerous reinforcement learning algorithms and environments are pouring out. Since each of the reinforcement learning environments represents a different problem, the algorithm must be written in accordance with the corresponding expression method. Therefore, it is necessary to know more about the reinforcement learning environment to be used. In this paper, several reinforcement learning environments that have been released for free are introduced with pictures and their features are listed. In particular, the multi-agent reinforcement learning environment is mainly described, and a comparative analysis table of the introduced reinforcement learning environments is presented at the end.

Statistics
Cite this article