Paper Title :A Glimpse at Reinforcement Learning Method

Author :Cetin Ozgur Baltaci, Serhan Ozdemir

Article Citation :Cetin Ozgur Baltaci ,Serhan Ozdemir , (2018 ) " A Glimpse at Reinforcement Learning Method " , International Journal of Advance Computational Engineering and Networking (IJACEN) , pp. 51-55, Volume-6, Issue-2

Abstract : This paper aims to describe the main theoretical details of Reinforcement Learning (RL). RL is the closest to human learning among all the other learning methods. Just as an infant learns by itself without any teacher supervision, an agent learns through RL. A rational agent makes inferences between the effects and their consequences. The term state is used to defined one of the possible positions where an agent is in the environment. A Reward maybe defined as a credit or a penalty for each and every state. Actions are possible operations decided by an agent. As an example, an agent navigates through a maze. Keywords - Bellman equation, Machine learning,Markov property,Reinforcement learning.

Type : Research paper

Published : Volume-6, Issue-2

	|		PDF	|	Viewed - 61	|	Published on 2018-04-12