JuryGCN

This is the implementation for KDD'22 paper: "JuryGCN: Quantifying Jackknife Uncertainty on Graph Convolutional Networks".

Requirements

-Python: 3.8
-Pytorch: 1.4.0
-numpy: 1.19.2
-scikit-learn: 1.1.3
-scipy: 1.10.1
-autograd: 1.5

Evaluation

UQ with the application to active node classification: python uq4al.py

Others

Please kindly cite our paper if you find it helpful to your research:

@inproceedings{kang2022jurygcn,
  title={JuryGCN: quantifying jackknife uncertainty on graph convolutional networks},
  author={Kang, Jian and Zhou, Qinghai and Tong, Hanghang},
  booktitle={Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining},
  pages={742--752},
  year={2022}
}