This Github repository contains the code for the Master Graduation Project 'Analyzing Policy Gradient Methods for Rapid Policy Transfer'.
It's implemented with PyTorch 1.6.0 and the higher higher-order-gradient library, and includes a functional form of common implementations of multiple RL policy-gradient algorithms.
It will soon be updated with some more methods and functionality, and a report of the algorithmic flow, results and other components.
I hope you have as much fun with it as I had building it!
Link to thesis: https://research.tue.nl/en/studentTheses/analyzing-policy-gradient-approaches-towards-rapid-policy-transfe