Some code related to my work

In this page I collected all the codes I shared alongside my papers. All these codes are written in Python or Julia (or both) and are either contained in a .zip folder or lead to a Github repository. For more details on the content of these codes, I invite to refer to the article in which the corresponding algorithm was introduced. If you make use of these codes, please cite us with bibliographic references indicated here below.

2024

• An embedding-based distance for temporal graphs

  In this repository you can find a Python implementation of our distance definition for dynamic graphs.

@misc{dallamico2024embeddingbased,
      title={An embedding-based distance for temporal graphs}, 
      author={Lorenzo Dall'Amico and Alain Barrat and Ciro Cattuto},
      year={2024},
      eprint={2401.12843},
      archivePrefix={arXiv},
      primaryClass={cs.SI}
}

2023

• Efficient distributed representations beyond negative sampling

  In this repository you can find a Python implementation of our algorithm to efficiently produce distributed representations of complex entities.

  @misc{dallamico2023efficient,
    title={Efficient distributed representations beyond negative sampling}, 
    author={Lorenzo Dall'Amico and Enrico Maria Belliardo},
    year={2023},
    eprint={2303.17475},
    archivePrefix={arXiv},
    primaryClass={cs.LG}
    }
  

  Python code

2021

• Nishimori meets Bethe: a spectral method for node classification in sparse weighted graphs (JSTAT)

  We share here a repository containing the code to reproduce the results of our work on weighted graphs with applications to cost-efficient clustering.

  @article{Dall_Amico_2021,
    doi = {10.1088/1742-5468/ac21d3},
    url = {https://doi.org/10.1088/1742-5468/ac21d3},
    year = 2021,
    month = {sep},
    publisher = {{IOP} Publishing},
    volume = {2021},
    number = {9},
    pages = {093405},
    author = {Lorenzo Dall'Amico and Romain Couillet and Nicolas Tremblay},
    title = {Nishimori meets Bethe: a spectral method for node classification in sparse weighted graphs},
    journal = {Journal of Statistical Mechanics: Theory and Experiment}
    }
  

  Julia zip, Julia repository

2020

• A unified framework for spectral clustering in sparse graphs Python code (JMLR)

  This is our main contribution to the field of spectral clustering for community detection. We share a Python implementation of the CoDeBetHe algorithm as well as a more efficient package written in Julia.

  @article{dall2021unified,
    author  = {Lorenzo Dall'Amico and Romain Couillet and Nicolas Tremblay},
    title   = {A Unified Framework for Spectral Clustering in Sparse Graphs},
    journal = {Journal of Machine Learning Research},
    year    = {2021},
    volume  = {22},
    number  = {217},
    pages   = {1-56},
    url     = {http://jmlr.org/papers/v22/20-261.html}
    }
  

  Python code, Julia repository, Documentation of the Julia package

• Community detection in sparse time-evolving graphs with a dynamical Bethe-Hessian (NeurIPS)

  Find here the Python implementation to reproduce the results of this paper in which we studied community detection in dynamical graphs

  @article{dall2020community,
    title={Community detection in sparse time-evolving graphs with a dynamical Bethe-Hessian},
    author={Dall'Amico, Lorenzo and Couillet, Romain and Tremblay, Nicolas},
    journal={Advances in Neural Information Processing Systems},
    volume={33},
    year={2020}
    }
  

  Python code

2019

• Revisiting the Bethe-Hessian: Improved Community Detection in Sparse Heterogeneous Graphs (NeurIPS)

  Find here the Python implementation to reproduce the results of this paper.

  @inproceedings{dall2019revisiting,
    title={Revisiting the Bethe-Hessian: improved community detection in sparse heterogeneous graphs},
    author={Dall'Amico, Lorenzo and Couillet, Romain and Tremblay, Nicolas},
    booktitle={Advances in Neural Information Processing Systems},
    pages={4039--4049},
    year={2019}
    }
  

  Python code