ToruOwO / marl-ae-comm

PyTorch implementation for all models and environments in the paper "Learning to Ground Multi-Agent Communication with Autoencoders"

Learning to Ground Multi-Agent Communication with Autoencoders

This repo contains the PyTorch implementation for all models and environments in the paper Learning to Ground Multi-Agent Communication with Autoencoders

by Toru Lin, Minyoung Huh, Chris Stauffer, Sernam Lim, and Phillip Isola.

Code layout

Please see each sub-directory for more details.

Directory	Detail
cifar-game	environment and models for training "CIFAR Game"
:-------------:	:-------------:
marl-grid/env	environments for training "FindGoal" and "RedBlueDoors"
marl-grid/find-goal	models for training "FindGoal"
marl-grid/red-blue-doors	models for training "RedBlueDoors"

Paper citation

If you used this code or found our work helpful, please consider citing:

@misc{lin2021learning,
      title={Learning to Ground Multi-Agent Communication with Autoencoders}, 
      author={Toru Lin and Minyoung Huh and Chris Stauffer and Ser-Nam Lim and Phillip Isola},
      year={2021},
      eprint={2110.15349},
      archivePrefix={arXiv},
      primaryClass={cs.LG}
}