Certification of Neural Network Reinforcement Learning Policies based on Randomized Smoothing

We partly modify the repository https://github.com/DLR-RM/rl-baselines3-zoo (in particular: https://github.com/DLR-RM/rl-baselines3-zoo/tree/b83409561347fe2035d1b4ce8d700ab3418c7246) to fulfill our Master thesis of RS certification of RL policies. All code credits belong to RL3Zoo team. Please email me if further discussion is needed.

Usage: please check out the arguments from enjoy_att.py and the command examples from the attack_ bash scripts. Besides, the original README from the RL3Zoo repository is also helpful.