# Introduction

This repository contains codes that are used for generating numerical results in the following paper: 

W. U. Mondal, V. Aggarwal, and S. V. Ukkusuri, "[On the Near-Optimality of Local Policies in Large Cooperative 
Multi-Agent Reinforcement Learning](https://openreview.net/pdf?id=t5HkgbxZp1)", Transactions on Machine Learning Research, 2022.


```
@article{
mondal2022on,
title={On the Near-Optimality of Local Policies in Large Cooperative Multi-Agent Reinforcement Learning},
author={Washim Uddin Mondal and Vaneet Aggarwal and Satish Ukkusuri},
journal={Transactions on Machine Learning Research},
year={2022},
url={https://openreview.net/forum?id=t5HkgbxZp1},
note={}
}
```

# Parameters

Various parameters used in the experiments can be found in [Scripts/Parameters.py](https://github.itap.purdue.edu/Clan-labs/NearOptimalLocalPolicy/blob/main/Scripts/Parameters.py) file.

# Results

Generated results will be stored in Results folder (will be created on the fly).
Some pre-generated results are available for display in the [Display](https://github.itap.purdue.edu/Clan-labs/NearOptimalLocalPolicy/tree/main/Display) folder. Specifically,
[Fig. 1a](https://github.itap.purdue.edu/Clan-labs/NearOptimalLocalPolicy/blob/main/Display/Fig1a.png) depicts the percentage error between the values generated by local and non-local policies in an N-agent system
as a function of N.

# Run Experiments

```
python3 Main.py
```

# Command Line Options

Various command line options are given below:

```
--train : if training is required from scratch, otherwise a pre-trained model will be used   
--minN : minimum value of N  
--numN : number of N values  
--divN : difference between two consecutive N values  
--maxSeed: number of random seeds 
```