Publications

Identifying counterfactual probabilities using bivariate distributions and uplift modeling
Théo Verhelst, Gianluca Bontempi

Preprint, Submitted to the 34th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, 2025
arXiv – 

@misc{verhelst2025identifying,
  title={Identifying counterfactual probabilities using bivariate distributions and uplift modeling}, 
  author={Verhelst, Th{\'e}o and Bontempi, Gianluca},
  year={2025},
  eprint={2512.08805},
  archivePrefix={arXiv},
  primaryClass={cs.LG}
}

Probability of collision in nonlinear dynamics by moment propagation
Théo Verhelst, Giacomo Acciarini, Dario Izzo, Francesco Biscani

Conference paper, 9th European Conference on Space Debris, 2025
Full paper – arXiv – 

@InProceedings{verhelst2025probability,
  title={Probability of collision in nonlinear dynamics by moment propagation}, 
  author={Verhelst, Th{\'e}o and Acciarini, Giacomo and Izzo, Dario and Biscani, Francesco},
  year={2025},
  volume = {9},
  number = {1},
  booktitle={9th European Conference on Space Debris},
  editor = {S. Lemmens and T. Flohrer and F. Schmitz},
  address = {Bonn, Germany},
  publisher = {ESA Space Debris Office},
  month = {April},
  url={https://conference.sdo.esoc.esa.int/proceedings/sdc9/paper/287/SDC9-paper287.pdf}, 
}

Causal and predictive modeling of customer churn: lessons learned from empirical and theoretical research
Théo Verhelst, supervised by Gianluca Bontempi

PhD thesis, Université Libre de Bruxelles, 2024
PDF – 

@thesis{verhelst2024causal,
  type = {phdthesis},
  title = {Causal and Predictive Modeling of Customer Churn: Lessons Learned from Empirical and Theoretical Research},
  shorttitle = {Causal and Predictive Modeling of Customer Churn},
  author = {Verhelst, Th{\'e}o},
  date = {2024-01},
  institution = {{Université libre de Bruxelles}},
  location = {{Bruxelles}},
  pagetotal = {225},
  url = {https://theoverhelst.com/assets/documents/theo_verhelst_phd_thesis.pdf},
  urldate = {2025-12-15}
}

Uplift vs. predictive modeling: a theoretical analysis
Théo Verhelst, Robin Petit, Wouter Verbeke, Gianluca Bontempi

Preprint, 2023
arXiv – 

@misc{verhelst2023uplift,
  title={Uplift vs. predictive modeling: a theoretical analysis},
  author={Verhelst, Th{\'e}o and Petit, Robin and Verbeke, Wouter and Bontempi, Gianluca},
  year={2023},
  eprint={2309.12036},
  archivePrefix={arXiv},
  primaryClass={cs.LG}
}

A churn prediction dataset from the telecom sector: a new benchmark for uplift modeling
Théo Verhelst, Denis Mercier, Jeevan Shrestha, Gianluca Bontempi

Conference paper, ECML PKDD 2023 Workshops - Workshop on Uplift Modeling and Causal Machine Learning for Operational Decision Making, 2023
Full paper – arXiv – 

@InProceedings{verhelst2025churn,
  title={A Churn Prediction Dataset from the Telecom Sector: A New Benchmark for Uplift Modeling},
  author={Verhelst, Th{\'e}o and Mercier, Denis and Shrestha, Jeevan and Bontempi, Gianluca},
  editor={Meo, Rosa and Silvestri, Fabrizio},
  year={2025},
  publisher={Springer Nature Switzerland},
  booktitle={Machine Learning and Principles and Practice of Knowledge Discovery in Databases},
  address={Cham},
  pages={292--299},
  isbn={978-3-031-74640-6},
  doi={10.1007/978-3-031-74640-6_21}
}

Partial counterfactual identification and uplift modeling: theoretical results and real-world assessment
Théo Verhelst, Denis Mercier, Jeevan Shrestha, Gianluca Bontempi

Journal paper, Machine Learning, 2023
Full paper – Read-only full text – arXiv – 

@article{verhelst2023partial,
  title={Partial counterfactual identification and uplift modeling: theoretical results and real-world assessment},
  author={Verhelst, Th{\'e}o and Mercier, Denis and Shrestha, Jeevan and Bontempi, Gianluca},
  journal={Machine Learning},
  pages={1--25},
  year={2023},
  publisher={Springer},
  doi={10.1007/s10994-023-06317-w}
}

Predicting reach to find persuadable customers: improving uplift models for churn prevention
Théo Verhelst, Denis Mercier, Jeevan Shrestha, Jean-Christophe Dewitte, Gianluca Bontempi

Conference paper, 24th International Conference on Discovery Science, 2021
Full paper – 

@InProceedings{verhelst2021predicting
  author={Verhelst, Th{\'e}o and Shrestha, Jeevan and Mercier, Denis and Dewitte, Jean-Christophe and Bontempi, Gianluca},
  editor={Soares, Carlos and Torgo, Luis},
  title={Predicting Reach to Find Persuadable Customers: Improving Uplift Models for Churn Prevention},
  booktitle="Discovery Science",
  year="2021",
  publisher="Springer International Publishing",
  address="Cham",
  pages="44--54",
  isbn="978-3-030-88942-5",
  doi={10.1007/978-3-030-88942-5_4}
}

Transfer learning strategies for credit card fraud detection
Bertrand Lebichot, Théo Verhelst, Yann-Aël Le Borgne, Liyun He-Guelton, Frédéric Oblé, Gianluca Bontempi

Journal paper, IEEE Access, 2021
Full paper – 

@article{lebichot2021transfer,
  author={Lebichot, Bertrand and Verhelst, Th{\'e}o and Le Borgne, Yann-Aël and He-Guelton, Liyun and Oblé, Frédéric and Bontempi, Gianluca},
  journal={IEEE Access},
  title={Transfer Learning Strategies for Credit Card Fraud Detection},
  year={2021},
  volume={9},
  number={},
  pages={114754-114766},
  doi={10.1109/ACCESS.2021.3104472}
}

Understanding telecom customer churn with machine learning: from prediction to causal inference
Théo Verhelst, Olivier Caelen, Jean-Christophe Dewitte, Bertrand Lebichot, Gianluca Bontempi

Conference paper, 31st Benelux Conference on Artificial Intelligence and the 28th Belgian Dutch Conference on Machine Learning, 2019
Full paper – 

@InProceedings{verhelst2021understanding,
  author={Verhelst, Th{\'e}o and Caelen, Olivier and Dewitte, Jean-Christophe and Lebichot, Bertrand and Bontempi, Gianluca},
  editor={Bogaerts, Bart and Bontempi, Gianluca and Geurts, Pierre and Harley, Nick and Lebichot, Bertrand and Lenaerts, Tom and Louppe, Gilles},
  title="Understanding Telecom Customer Churn with Machine Learning: From Prediction to Causal Inference",
  booktitle="Artificial Intelligence and Machine Learning",
  year="2020",
  publisher="Springer International Publishing",
  address="Cham",
  pages="182--200",
  isbn="978-3-030-65154-1",
  doi={10.1007/978-3-030-65154-1_11}
}

Churn Prediction and Causal Analysis on Telecom Customer Data
Théo Verhelst, supervised by Gianluca Bontempi

Master thesis, Université Libre de Bruxelles, 2018
PDF – 

@thesis{verhelst2018churn,
  type = {mathesis},
  title = {Churn Prediction and Causal Analysis on Telecom Customer Data},
  author = {Verhelst, Th{\'e}o},
  date = {2018},
  institution = {{Université libre de Bruxelles}},
  location = {{Brussels, Belgium}},
  url = {https://theoverhelst.com/assets/documents/theo_verhelst_master_thesis.pdf},
  urldate = {2025-12-15},
  pagetotal = {83}
}