An investigation of the standard Q-Learning approach as well as the SARSA control policy and Double Q-Learning for the control problems Cart Pole and Mountain Car. To understand the variations in performance across these implementations and environments.
Developed for Python 3.8
Report detailing production is available on GitHub: podit/reportmentLearning
pip install pipenv
pipenv install
Use an experiment template as a guide to get the plot desired and enter experimental values.
Use virtual environment to run the experimental script from the top level directory:
pipenv shell
python experimentName.py