/RawlsGCN

Implementations of 'RawlsGCN: Towards Rawlsian Difference Principle on Graph Convolutional Network', WWW'22

Primary LanguagePythonMIT LicenseMIT

RawlsGCN: Towards Rawlsian Difference Principle on Graph Convolutional Network

Implementations of 'RawlsGCN: Towards Rawlsian Difference Principle on Graph Convolutional Network', WWW'22

Requirements

Main dependency: pytorch

Tested under python 3.8, pytorch 1.9

Run

RawlsGCN-Graph:

python train.py --model rawlsgcn_graph

RawlsGCN-Grad:

python train.py --model rawlsgcn_grad

Citation

If you find this repository useful, please kindly cite the following paper:

@inproceedings{kang2022rawlsgcn,
  title={Rawlsgcn: Towards rawlsian difference principle on graph convolutional network},
  author={Kang, Jian and Zhu, Yan and Xia, Yinglong and Luo, Jiebo and Tong, Hanghang},
  booktitle={Proceedings of the ACM Web Conference 2022},
  pages={1214--1225},
  year={2022}
}