SLOG: An Inductive Spectral Graph Neural Network Beyond Polynomial Filter

Welcome to the official code repository for SLOG: An Inductive Spectral Graph Neural Network Beyond Polynomial Filter, accepted at ICML 2024.

Installation

The code is implemented in PyTorch. To install the required packages, run:

pip install -r requirements.txt

To modify the configurations of the experiments, please refer to config.py. You can change the dataset/model/optimization and other hyperparameters in the configuration file. To run the code, please execute:

python main.py

Contact

For immediate queries or further information, please open an issue or contact xuhb20@mails.tsinghua.edu.cn or yucheny5@illinois.edu.

Citation

If you find our work useful in your research, please consider citing our paper:

@inproceedings{xuslog,
  title={SLOG: An Inductive Spectral Graph Neural Network Beyond Polynomial Filter},
  author={Xu, Haobo and Yan, Yuchen and Wang, Dingsu and Xu, Zhe and Zeng, Zhichen and Abdelzaher, Tarek F and Han, Jiawei and Tong, Hanghang},
  booktitle={Forty-first International Conference on Machine Learning}
}