A PettingZoo Gym-based environment for microelectrode-dot-array (MEDA) biochips.
This environment simulates the dynamic environment of a MEDA biochip.
For furhter information, please reference our ICML'21 paper named "Parallel Droplet Control in MEDA Biochips using Ｍulti-Agent Reinforcement Learning".
First, install all the required libraries when you run meda_env/train.py. Then you can find the training result under the 'log' directory.
To cite this work in publications:
@inproceedings{liang2021parallel,
title = {Parallel Droplet Control in MEDA Biochips using Ｍulti-Agent Reinforcement Learning},
author = {Liang, Tung-Che and Zhou, Jin and Chan, Yun-Sheng and Ho, Tsung-Yi, and Chakrabarty, Krishnendu and Lee, Chen-Yi},
booktitle={International Conference on Machine Learning (ICML)},
year={2021}
}