Official Implementation of Half-Hop: A graph upsampling approach for slowing down message passing (ICML 2023)
Half-Hop is plug-and-play, and works with a wide range of datasets, architectures, and learning objectives!
Example usage:
from halfhop import HalfHop
# apply augmentation
transform = HalfHop(alpha=0.5)
data = transform(data)
# feedforward
y = model(data)
# get rid of slow nodes
y = y[~data.slow_node_mask]If you find the code useful for your research, please consider citing our work:
@article{azabou2023half,
title={Half-Hop: A graph upsampling approach for slowing down message passing},
author={Azabou, Mehdi and Ganesh, Venkataramana and Thakoor, Shantanu and Lin, Chi-Heng and Sathidevi, Lakshmi and Liu, Ran and Valko, Michal and Veli{\v{c}}kovi{\'c}, Petar and Dyer, Eva L},
journal={Proceedings of the International Conference on Machine Learning (ICML)}
year={2023}
}