This implementation is based on Deep Inverse Q-learning with Constraints
Arxiv : https://arxiv.org/abs/2008.01712
Blog Post: http://nrprojects.cs.uni-freiburg.de/foundations.html#inverse
Using the Inverse Q-Learning on vector states Evaluation on the LunarLander enviroment from open ai gym The expert samples are generated for an policy that solved the problem with an reward at least 200
- Clone the repository and navigate to the downloaded folder.
git clone https://github.com/ChrisProgramming2018/IQL_Lunar_Lander.git
cd IQL_Lunar_Lander
- Create a conda enviroment
conda create -n iql python=3.7
- Activate the Enviroment
conda activate iql
- Install all dependencies, use the requirements.txt file
pip3 install -r requirements.txt