https://github.com/sunfanyunn/InfoGraph

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

graph-level-representation graph-neural-networks infograph mutual-information-neural-estimator

Last synced: about 2 months ago
JSON representation

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

• Host: GitHub
• URL: https://github.com/sunfanyunn/InfoGraph
• Owner: sunfanyunn
• Created: 2018-11-23T05:54:36.000Z (about 6 years ago)
• Default Branch: master
• Last Pushed: 2023-12-27T00:40:59.000Z (about 1 year ago)
• Last Synced: 2024-08-14T20:08:30.175Z (5 months ago)
• Topics: graph-level-representation, graph-neural-networks, infograph, mutual-information-neural-estimator
• Language: Python
• Homepage: https://openreview.net/forum?id=r1lfF2NYvH
• Size: 131 MB
• Stars: 312
• Watchers: 7
• Forks: 45
• Open Issues: 3
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

• awesome-graph-based-semi-supervised-learning - code

README

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

#### Authors: [Fan-yun Sun](https://fanyun-sun.github.io/fanyun-sun.github.io), [Jordan Hoffman](https://jhoffmann.org/), [Vikas Verma](http://vikasverma1077.github.io/), [Jian Tang](https://jian-tang.com/)

#### [Link to Paper](https://openreview.net/forum?id=r1lfF2NYvH)

Tested on pytorch 1.6.0 and [pytorch\_geometric](https://github.com/rusty1s/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 codebase with `pytorch_geometric>=1.6.1`, you may obtain results different from those reported on the paper.

### Citation

Please cite [our paper](https://openreview.net/pdf?id=r1lfF2NYvH) 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}

}

```

We thank the following work: [Deep InfoMax](https://github.com/rdevon/DIM)