BVAT (ICML2019 workshop paper)

Code for "Batch Virtual Adversarial Training for Graph Convolutional Networks" which is based on the original implementation of GCN.

Requirements

• tensorflow (>0.12)
• networkx

Run the codes

We provide two adversarial training algorithms (SBVAT and OBVAT). Please refer to our paper for the details. Typically, you can run the algorithms by:

cd obvat
python train.py

Cite

Please cite our paper if you use this code in your own work:

@article{deng2019batch,
  title={Batch Virtual Adversarial Training for Graph Convolutional Networks},
  author={Deng, Zhijie and Dong, Yinpeng and Zhu, Jun},
  journal={arXiv preprint arXiv:1902.09192},
  year={2019}
}