Maximum Entropy Inverse Reinforcement Learning

Course Project for the "Inverse Reinforcement Learning" Seminar at the University of Stuttgart.

Dependencies

The requirements are:

• numpy
• matplotlib
• gym
• ipython
• jupyter

Requirements can be installed as follows:

pip install -r requirements.txt

Submission notebook

The notebook submission file is named maximum_entropy_irl.ipynb.
It was tested using juypter notebook on Ubuntu 20.04.

Notes

• After the presentation, some changes were made (listed in the notebook) to address the results from the discussion
• Slides and Notebook no longer match exactly