Gradient Gating for Deep Multi-Rate Learning on Graphs

This repository contains the implementation to reproduce the numerical experiments of the ICLR 2023 paper Gradient Gating for Deep Multi-Rate Learning on Graphs

Requirements

Main dependencies (with python >= 3.7):
torch==1.9.0
torch-cluster==1.5.9
torch-geometric==2.0.3
torch-scatter==2.0.9
torch-sparse==0.6.12
torch-spline-conv==1.2.1

Commands to install all the dependencies in a new conda environment
(python 3.7 and cuda 10.2 -- for other cuda versions change accordingly)

conda create --name gradientgating python=3.7
conda activate gradientgating

pip install torch==1.9.0

pip install torch-scatter -f https://pytorch-geometric.com/whl/torch-1.9.0+cu102.html
pip install torch-sparse -f https://pytorch-geometric.com/whl/torch-1.9.0+cu102.html
pip install torch-cluster -f https://pytorch-geometric.com/whl/torch-1.9.0+cu102.html
pip install torch-spline-conv -f https://pytorch-geometric.com/whl/torch-1.9.0+cu102.html
pip install torch-geometric
pip install scipy
pip install numpy

Citation

If you found our work useful in your research, please cite our paper at:

@inproceedings{rusch2022gradient,
  title={Gradient Gating for Deep Multi-Rate Learning on Graphs},
  author={Rusch, T Konstantin and Chamberlain, Benjamin P and Mahoney, Michael W and Bronstein, Michael M and Mishra, Siddhartha},
  booktitle={International Conference on Learning Representations},
  year={2023}
}

(Also consider starring the project on GitHub.)