Q-Learning using Neural Networks in OpenAI Gym

Overview

Q-Learning is one of the Reinforcement Learning Algorithm. In this tutorial we are going to deploy the variant of Q-Learning using neural networks for mapping state space to action space experimented on OpenAI Gym "FrozenLake" environment.

You can find an excellent tutorial on Q-learning using Neural Networks by Arthur Juliani here.

Dependencies

• Jupyter Notebook
• Numpy
• OpenAI Gym (https://gym.openai.com/docs/)
• Tensorflow (https://www.tensorflow.org/install/)

Credits

Most of the conceptual and programmatic understanding is borrowed from the Reinforcement Learning Series by Arthur Juliani here.