Understanding Oversmoothing in Diffusion-Based GNNs From the Perspective of Operator Semigroup Theory

This repository contains the code for "Understanding Oversmoothing in Diffusion-Based GNNs From the Perspective of Operator Semigroup Theory", which is primarily based on the GREAD framework.

Usage

To use the implementation:

1. Clone the original GREAD GitHub Repository:

   git clone https://github.com/jeongwhanchoi/GREAD.git

2. Replace the function_gread.py and gread_params.py files in the original repository with the function_gread.py and gread_params.py provided in this repository
3. Run the program and specify the desired breaking term during runtime. For instance: To run each experiment, navigate into src. Then, run the following command:

   python run_GNN.py --dataset=texas --use_best_params
   

Citation

@article{zhao2024understanding,
  title={Understanding Oversmoothing in Diffusion-Based GNNs From the Perspective of Operator Semigroup Theory},
  author={Zhao, Weichen and Wang, Chenguang and Wang, Xinyan and Han, Congying and Guo, Tiande and Yu, Tianshu},
  journal={arXiv preprint arXiv:2402.15326},
  year={2024}
}