We introduce Sorrel (https://github.com/social-ai-uoft/sorrel), a simple Python interface for generating and testing new multi-agent reinforcement learning environments. This interface places a high degree of emphasis on simplicity and accessibility, and uses a more psychologically intuitive structure for the basic agent-environment loop, making it a useful tool for social scientists to investigate how learning and social interaction leads to the development and change of group dynamics. In this short paper, we outline the basic design philosophy and features of Sorrel.
@article{arxiv.2506.00228,
title = {Sorrel: A simple and flexible framework for multi-agent reinforcement learning},
author = {Rebekah A. Gelpí and Yibing Ju and Ethan C. Jackson and Yikai Tang and Shon Verch and Claas Voelcker and William A. Cunningham},
journal= {arXiv preprint arXiv:2506.00228},
year = {2025}
}