network-deconfounder-wsdm20

Code for the WSDM '20 paper, Learning Individual Causal Effects from Networked Observational Data.

First Work on Causal Effect Estimation Based on Graph Convolutional Neural Networks

For now, please cite the WSDM version if you find this paper/repository is helpful.

@inproceedings{guo2020learning,
  title={Learning Individual Causal Effects from Networked Observational Data},
  author={Guo, Ruocheng and Li, Jundong and Liu, Huan},
  booktitle={Proceedings of the 13th International Conference on Web Search and Data Mining},
  pages={232--240},
  year={2020}
}

Overview of the Network Deconfounder

Dependencies

Tested on Ubuntu 18.04

Python 3.6
Pytorch 1.2.0
Scipy 1.3.1
Numpy 1.17.2
Pandas 0.25.1

Datasets

Datasets used in this paper can be found in ./datasets

Running the experiment

On a linux system, you can run the bash script, for example

bash run_for_share.sh --dataset BlogCatalog

Then results would be added to the end of the corresponding csv files in ./new_results/Blogcatalog or ./new_results/Flickr

To calculate the mean of the results from 10 simulations, you can use the python script

python res_mean.py