Contents:
This is a repository for "Multi-Agent Actor-Critic Network-Based Incentive Mechanism for Mobile Crowdsensing in Industrial System".
- Python 3.7.3
- Numpy
- cvxpy
- matplotlib
- pyyaml
- tensorboardX
- pytorch
You can install all the dependencies with pip:
cd environment
pip install -r requirments.txt
There is a simple script with which you will begin the training process. A sample command to launch training is
python train.py
Moreover, the settings for training are decribed in platform-config.yaml at \code\environment directory. And if needed, you can modify the script to satisfy other purposes.
If this repository do some help, please consider to cite the paper "Multi-Agent Actor-Critic Network-Based Incentive Mechanism for Mobile Crowdsensing in Industrial System".
B. Gu, X. Yang, Z. Lin, W. Hu, M. Alazab and R. Kharel, "Multi-Agent Actor-Critic Network-based Incentive Mechanism for Mobile Crowdsensing in Industrial Systems," in IEEE Transactions on Industrial Informatics, doi: 10.1109/TII.2020.3024611.