Review on Reinforcement Learning in CartPole Game

Last recent years, reinforcement learning has been one of the machine learning approaches commonly used in many fields. Reinforcement learning applications can be in games, resource management, personalized recommendations, and robotics. The applications of RL implementation in the game increase rapidly. Therefore, ensuring the best performance of RL applications in games is one of the challenges that should be considered. The reinforcement learning methods can be used to achieve the highest scores in the game in the minimum time possible to win. This work presents an overview of RL with focusing on Q-learning and State Action Reward State Action (SARSA) models. Also, this work describes RL on CartPole Game and applies an experiment to measure and compare the implementation performance of Q-learning and SARSA in CartPole Game.