Learning Laplacian Positional Encodings for Heterophilous Graphs

[AISTATS 2025] Learning Laplacian Positional Encodings for Heterophilous Graphs

Instructions for Reproducibility

1. In the generation directory, use the following command to download and preprocess datasets from PyTorch Geometric:

   source generate_benchmarks.sh

2. Once the datasets are downloaded and preprocessed, in the root directory, use the following command to train models and save results to the results directory.

   source search.sh

3. After training the models and saving the results, load the results in the load_results.ipynb notebook located in the results directory.

Citation

If you find this work useful, please cite our paper:

@inproceedings{ItoKW25llpe,
  author       = {Michael Ito and Jiong Zhu and Dexiong Chen and Danai Koutra and Jenna Wiens},
  title        = {Learning Laplacian Positional Encodings for Heterophilous Graphs},
  booktitle    = {International Conference on Artificial Intelligence and Statistics},
  publisher    = {PMLR},
  year         = {2025},
}