GNN-Generalization-Fairness

This repo provides the official implementations for the experiments described in the following paper:

Subgroup Generalization and Fairness of Graph Neural Networks

Jiaqi Ma*, Junwei Deng*, and Qiaozhu Mei. NeurIPS 2021.

(*: equal constribution)

Requirements

plantoid-experiments

• dgl 0.4.3
• torch 1.8.0
• networkx 2.4
• numpy 1.17.3

ogb-experiments

• torch 1.8.2
• torch-geometric
• networkx 2.4
• ogb 1.3.2
• numpy 1.17.3

Run the code

Please refer to the readme in each subfolder.

Cite

@article{DBLP:journals/corr/abs-2106-15535,
  author    = {Jiaqi Ma, Junwei Deng, Qiaozhu Mei},
  title     = {Subgroup Generalization and Fairness of Graph Neural Networks},
  journal   = {CoRR},
  volume    = {abs/2106.15535},
  year      = {2021},
  url       = {https://arxiv.org/abs/2106.15535},
  eprinttype = {arXiv},
  eprint    = {2106.15535},
  timestamp = {Mon, 05 Jul 2021 15:15:50 +0200},
  biburl    = {https://dblp.org/rec/journals/corr/abs-2106-15535.bib},
  bibsource = {dblp computer science bibliography, https://dblp.org}
}