In this repo you can find the code for the paper "A Variational Approach to Privacy and Fairness" from the 2nd Privacy Preserving Artificial Intelligence (PPAI) Workshop of the AAAI Conference on Artificial Intelligence.
A guide on how to run the code can be found on this notebook.
The code has been tested on Python 3.6.8 and the following packages. It will probably work with older packages, though.
- torch 1.2.0+cpu
- torchvision 0.4.0+cpu
- matplotlib 3.0.2
- progressbar2 3.43.1
- scikit-learn 0.22.1
- numpy 1.18.4
- pandas 1.0.3
- progressbar2 3.39.2
- umap
- numba 0.48.0
In order to install the requirements you can just write pip3 install -r requirements.txt.