Generalized Demographic Parity for Group Fairness

Codes for [ICLR'22] Generalized Demographic Parity for Group Fairness.

Zhimeng Jiang, Xiaotian Han, Chao Fan, Fan Yang, Ali Mostafavi, Xia Hu

GDP Overview

Implementations of GDP

Set Environment

conda env create -f environment.yml

Dataset

1. Tabular data: Adults and Crimes dataset can be downloaded from (https://archive.ics.uci.edu/ml/datasets/adult) and (https://archive.ics.uci.edu/ml/datasets/communities+and+crime)

2. Graph data: Pokec_z and Pokec_n can be downloaded from (https://github.com/EnyanDai/FairGNN/tree/main/dataset/pokec) as region_job.xxx and region_job_2.xxx, respectively. They are sampled from soc_Pokec.

Reproduce the results

To reproduce the performance reported in the paper, you can run the bash files in folders tabular\,graph\ and comp\.

cd tabular
bash run_dnn.sh
bash run_dnn_adv.sh

Citation

@inproceedings{jiang2022generalized,
  title={Generalized Demographic Parity for Group Fairness},
  author={Jiang, Zhimeng and Han, Xiaotian and Fan, Chao and Yang, Fan and Mostafavi, Ali and Hu, Xia},
  booktitle={International Conference on Learning Representations},
  year={2022}
}

License

MIT