Mix-ME: Quality-Diversity for Multi-Agent Learning

This repository is part of a UCL Master's project and has been adapted into a workshop paper presented at the ALOE Workshop at NeurIPS 2023.

Features

• Environment Renders: Stores environment renders as HTML files in the trajectories folder.
• Knowledge Distillation: Implements knowledge distillation into a single network through imitation learning.