A Survey of Markov Model in Reinforcement Learning

Tianhan Gao, Baicheng Chen, Qingwei Mi

Abstract: There was a famous mathematician from Russian in the early 20th century whose name is Andrey Andreyevich Markov. After he proposed the Markov process, the theories have been effectively developed in the past years and have been widely adopted in different fields. This paper both summarizes and does incomplete statistics on Markov models which are related to reinforcement learning. It will help the readers to quickly clarify the relationship between these theories and have an overall understanding of them.