Trained a Reinforcement Learning agent on Mountain Car environment
Published:
Trained an agent in a way that mountain car can navigate itself to the destination by altering velocity and position. Q-learning algorithm was used and environment was taken from OpenAI’s gym.
Data: OpenAI Gym
Algorithm: Q-Learning
Major tool(s): Numpy, Matplotlib
Code link
Agent performs the task successfully as shown below:
Reward metrics:
