Permalink
Sep 7, 2022
Sep 14, 2022
Sep 14, 2022
Sep 14, 2022
Sep 7, 2022
Sep 14, 2022
Sep 7, 2022
Sep 14, 2022
Sep 7, 2022
Newer
100644
48 lines (35 sloc)
1.73 KB
1
# Introduction
2
3
This repository contains codes that are used for generating numerical results in the following paper:
4
5
W. U. Mondal, V. Aggarwal, and S. V. Ukkusuri, "[On the Near-Optimality of Local Policies in Large Cooperative
6
Multi-Agent Reinforcement Learning](https://openreview.net/pdf?id=t5HkgbxZp1)", Transactions on Machine Learning Research, 2022.
7
8
9
```
10
@article{
11
mondal2022on,
12
title={On the Near-Optimality of Local Policies in Large Cooperative Multi-Agent Reinforcement Learning},
13
author={Washim Uddin Mondal and Vaneet Aggarwal and Satish Ukkusuri},
14
journal={Transactions on Machine Learning Research},
15
year={2022},
16
url={https://openreview.net/forum?id=t5HkgbxZp1},
17
note={}
18
}
19
```
20
21
# Parameters
22
23
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.
24
25
# Results
26
27
Generated results will be stored in Results folder (will be created on the fly).
28
Some pre-generated results are available for display in the [Display](https://github.itap.purdue.edu/Clan-labs/NearOptimalLocalPolicy/tree/main/Display) folder. Specifically,
29
[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
30
as a function of N.
31
32
# Run Experiments
33
34
```
35
python3 Main.py
36
```
37
38
# Command Line Options
39
40
Various command line options are given below:
41
42
```
43
--train : if training is required from scratch, otherwise a pre-trained model will be used
44
--minN : minimum value of N
45
--numN : number of N values
46
--divN : difference between two consecutive N values
47
--maxSeed: number of random seeds
48
```