This repository contains the code for the paper Fairness through Aleatoric Uncertainty. This paper will be presented at the 32nd ACM International Conference on Information and Knowledge Management (CIKM).
Desired pre-requisites are mentioned in requirements.txt. Please add additional requirements if needed.
Please use the script run_experiments.py. The results will be stored in the compiled_results folder.
Please use the fair-mixup papers code and run data_processing.py to generate the train/test splits. Change the CELEBA_PATH variable in utils/jax/models/image.py to point to the celeba directory generated with the splits.
Image experiments can be executed using run_image_experiments.py.
If you find this code useful, please cite our paper:
@article{tahir2023fairness,
title={Fairness through Aleatoric Uncertainty},
author={Tahir, Anique and Cheng, Lu and Liu, Huan},
journal={arXiv preprint arXiv:2304.03646},
year={2023}
}