This repository contains the official code implementation for the paper Cell Attention Networks.

Cell Attention Networks propose a novel message-passing scheme for graph neural networks (GNNs) that lifts node feature vectors into a higher-dimensional space called Cellular Attention Networks. The information exchange between edges is weighted by learnable attention coefficients, which enhances the model's expressiveness and generalization.

• Requirements
• Installation
• Running Experiments
• Examples
• References
• Citation

• Python 3.7+
• PyTorch 1.9+
• torchvision 0.10+
• torch-geometric 2.0+
• numpy 1.20+
• tqdm 4.62+

To install the required dependencies, run the following command:

pip install -r requirements.txt

python ./exp/run_tu.py

sh ./exp/run_table.sh

If you find this work useful, please consider citing the paper:

@misc{giusti2022cell,
      title={Cell Attention Networks}, 
      author={Lorenzo Giusti and Claudio Battiloro and Lucia Testa and Paolo Di Lorenzo and Stefania Sardellitti and Sergio Barbarossa},
      year={2022},
      eprint={2209.08179},
      archivePrefix={arXiv},
      primaryClass={cs.LG}
}