https://github.com/papachristoumarios/dp-distributed-estimation

Supplementary code for "Differentially Private Distributed Estimation and Learning"
https://github.com/papachristoumarios/dp-distributed-estimation

Last synced: 5 months ago
JSON representation

Supplementary code for "Differentially Private Distributed Estimation and Learning"

• Host: GitHub
• URL: https://github.com/papachristoumarios/dp-distributed-estimation
• Owner: papachristoumarios
• License: mit
• Created: 2023-05-04T21:26:31.000Z (over 2 years ago)
• Default Branch: main
• Last Pushed: 2024-03-27T14:29:01.000Z (over 1 year ago)
• Last Synced: 2025-04-12T04:53:06.252Z (6 months ago)
• Language: Python
• Homepage: https://arxiv.org/abs/2306.15865
• Size: 30.2 MB
• Stars: 4
• Watchers: 3
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE
  • Citation: CITATION.cff

Awesome Lists containing this project

README

          
# dp-distributed-estimation

Supplementary code for the paper _Differentially Private Distributed Estimation and Learning_. 

## Installation and Usage

To install the required packages use `pip install requirements.txt`. To run the distributed estimation algorithms use

```bash

    python mean_estimation.py --name {name} --eps 1 --delta 0.01 

```

The following datasets are available:

 * `us_power_grid`: US Power Grid Network

 * `germany_consumption`: German Households Consumption Dataset

 * `ieee_33_bus`: IEEE 33 Bus network

For more options run `python mean_estimation.py --help`. 

## Reproducing Experiments

To reproduce the experiments of the paper run the `run_all_experiments.sh` script as 

```bash

bash run_all_experiments.sh

```

The resulting figures will be saved in the `figures` directory.