laonahongchen / soped-sdoped

Repo for my undergrad_thesis

undergrad_thesis

Repo for my undergraduate thesis project: A Research on the Method of Automatic Curriculum Learning for Multi-agent Reinforcement Learning.

This is a method for better automatic curriculum learning for RL training environment design. You can see the SOPED_SDOPED.pdf to get a brief idea about this work. If you are further interested, please contact me to get the paper or have a discussion.(Chinese only, because this is my undergraduate thesis LOL, but I would love to have a discussion about it if you want.).

Code largely based on PAIRED by Google.