Phantom

A Multi-agent reinforcement-learning simulator framework.
Explore the docs »
Report Bug · Request Feature

Table of Contents

1. About Phantom
2. Installation
3. Getting Started
4. Contributing
5. Citing Phantom
6. License

About Phantom

Phantom is a multi-agent reinforcement-learning simulator built on top of RLlib.

(back to top)

Installation

Prerequisites

The main requirements for installing Phantom are a modern Python installation (3.8 minimum) and access to the pip Python package manager.

A list of Python packages required by Phantom is given in the requirements.txt files in each respective directory. The required packages can be installed by running:

make install-deps

Phantom

Phantom can be installed as libraries with the command::

make install

To use the network plotting feature for Tensorboard the following additional packages are required:

• matplotlib
• networkx

(back to top)

Getting Started

With Phantom installed you can run the provided supply-chain sample experiment with the command:

phantom examples/environments/supply-chain/supply-chain.py

Change the script for any of the other provided experiments in the examples directory.

(back to top)

Contributing

Thank you for your interest in contributing to Phantom!

We invite you to contribute enhancements. Upon review you will be required to complete the Contributor License Agreement (CLA) before we are able to merge.

If you have any questions about the contribution process, please feel free to send an email to open_source@jpmorgan.com.

(back to top)

Citing Phantom

Find the paper on Arxiv Phantom -- An RL-driven framework for agent-based modeling of complex economic systems and markets or use the following BibTeX:

@inproceedings{ardon2023phantom,
  title={Phantom-A RL-driven Multi-Agent Framework to Model Complex Systems},
  author={Ardon, Leo and Vann, Jared and Garg, Deepeka and Spooner, Thomas and Ganesh, Sumitra},
  booktitle={Proceedings of the 2023 International Conference on Autonomous Agents and Multiagent Systems},
  pages={2742--2744},
  year={2023},
  keywords = {Artificial Intelligence, Multiagent Systems, Reinforcement Learning}
}

(back to top)

License

Distributed under the Apache 2.0 License. See LICENSE for more information.

(back to top)