Preliminary research investigating adversarial policies: given a victim policy in a multi-agent system, find a policy which will break the victim.

This codebase assumes Python 3.6. Install the requirements in requirements-build.txt before those in requirements.txt. Anaconda users can install directly with conda env create -f environment.yml.

Then install one of requirements-aprl.txt or requirements-modelfree.txt depending on which experiments you need to run (unfortunately these codebases depend on different MuJoCo and Gym versions.)

Finally, pip install -e . to install an editable version of the package.

Please run the ci/code_checks.sh before committing. This runs several linting steps. These are also run as a continuous integration check.

I like to use Git commit hooks to prevent bad commits from happening in the first place:

ln -s ../../ci/code_checks.sh .git/hooks/pre-commit