Get Started

The Python version used is 3.10.6.

This repository contains the code from our efforts to formalize a scalable theory of mind in multi-agent systems. We published the design motivation for the architecture used here in a preliminary work at the Agentic Markets workshop at ICML 2024, titled Scalable Approaches for a Theory of Many Minds.

git clone git@github.com:amimem/ToMM.git
cd ToMM

Create a virtual environment and install the requirements.

python -m venv venv
source venv/bin/activate
pip install -r requirements.txt

python mlp_match.py ARGUMENTS

If you are using Slurm, create a symlink to the scratch folder for the logs.

ln -s $SCRATCH scratch
sbatch scripts/slurm.sh