Unity ML-Agents project that acts as a labyrinth environment for compositional reinforcement learning algorithms. This repository contains the Unity project files necessary to construct the environment. Python code that interacts with this environment and trains compositional reinforcement learning systems may be found at the following github repository.
- Begin by installing the Unity Hub. Here is the download link.
- Next use the Unity Hub to install the appropriate version of Unity (This project was built using Unity 2020.3.20f1).
- Finally, after cloning this repository locally, drag the entire folder structure into the project list on Unity Hub. Use the Unity Hub to open the newly created project in order to get started.
To cite this project, please use:
@inproceedings{neary2022verifiable,
title={Verifiable and compositional reinforcement learning systems},
author={Neary, Cyrus and Verginis, Christos and Cubuktepe, Murat and Topcu, Ufuk},
booktitle={Proceedings of the International Conference on Automated Planning and Scheduling},
volume={32},
pages={615--623},
year={2022}
}