This project is aimed to solve the MountainCar-v0 environment in OpenAI Gym. The task is to learn to ride a car to the top of mountain against gravity by gaining adequate momentum.
The observation space is the position and velocity of the car at any moment. The continuous state space is discretised into blocks of 10x10.
The agent is rewarded -1 for every second of the episode! The episode ends when the car reaches the top of the mountain.