Creating a reinforcement learning model using Q-learning to train AI car to climb up the mountain
To understand reinforcement Q-learning and how to implement the same. MeanWhile, understanding how to use OpenAI-GYM library.
To use Mountain Car game from OpenAI-GYM library and implement the Q-learning reinforcement algorithm to play the game.
Using Greedy Epsilon Value algorithm, Q-value Action-value function, Mountain Car model of OpenAI-Gym to produce a reinforced model.
Mountain Car Model link: https://www.gymlibrary.dev/environments/classic_control/mountain_car/
Image of training around 2200 with decreasing Epsilon value.
Image of training around 10000 with decreasing Epsilon value.
Image of training around 37000 with constant Epsilon value or exploration ended.
Rewards Earned: -108.0
Rewards Earned: -143.0
Rewards Earned: -190.0
Rewards Earned: -107.0