Skip to content

Official code for "InfoGraph: Unsupervised and Semi-supervised Graph-Level Representation Learning via Mutual Information Maximization" (ICLR 2020, spotlight)

Notifications You must be signed in to change notification settings

tonydavis629/InfoGraph

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

73 Commits
 
 
 
 
 
 

Repository files navigation

InfoGraph: Unsupervised and Semi-supervised Graph-Level Representation Learning via Mutual Information Maximization

Tested on pytorch 1.6.0 and pytorch_geometric 1.6.1

Experiments reported on the paper are conducted in 2019 with pytorch_geometric==1.3.1. Note that the code regarding of QM9 dataset in pytorch_geometric has been changed since then. Thus, if you run this repo with pytorch_geometric>=1.6.1, you may obtain results differ from those reported on the paper.

Code regarding mutual information maximization is partially referenced from: https://github.com/rdevon/DIM

Cite

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

@inproceedings{sun2019infograph,
  title={InfoGraph: Unsupervised and Semi-supervised Graph-Level Representation Learning via Mutual Information Maximization},
  author={Sun, Fan-Yun and Hoffman, Jordan and Verma, Vikas and Tang, Jian},
  booktitle={International Conference on Learning Representations},
  year={2019}
}

About

Official code for "InfoGraph: Unsupervised and Semi-supervised Graph-Level Representation Learning via Mutual Information Maximization" (ICLR 2020, spotlight)

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%