profile_picture
Alexandre Forel
Postdoctoral Fellow, Polytechnique Montreal
alexandre.forel@polymtl.ca

I am a Researcher (previously Postdoctoral Researcher) at Polytechnique Montreal working with Prof. Thibaut Vidal. Before that, I obtained my Ph.D. at the Technical University of Munich under the supervision of Prof. Martin Grunow. I have a Master of Science in Engineering from the KTH Royal Institue of Technology in Stockholm and a Master of Engineering with a major in Control Theorey from Grenoble-INP ENSE3.

My research lies at the intersection of Operations Research and Machine Learning. I have a deep interest in data-driven optimization: how to transform data into decisions. I have worked on forecast evolution models, contextual optimization, and structured learning. Recently, I have also studied the other direction: explaining decisions by linking them back to the data.

News

  • I presented our work on the Differentiable feasibility pump at the Dagstuhl Seminar on Machine-Learning Augmented Combinatorial Optimization. The preprint is now available, October 2024.
  • DistrictNet: Decision-aware learning for geographical districting is accepted at NeurIPS 2024! Cheikh will be presenting the poster in Vancouver, August 2024.
  • It was a great pleasure to present our work on Functionally-identical pruning of tree ensembles at ISMP 2024. The paper is availble on arXiv, July 2024.
  • CF-OPT: Counterfactual explanations for data-driven optimization pipelines is accepted at ICML 2024! Germain will be presenting the poster in Vienna, June 2024.
  • Our work on Robust counterfactuals for randomized ensembles is accepted at CPAIOR 2024! I will be presenting it in Uppsala, Sweden, April 2024.

Publications

DistrictNet: Decision-aware learning for geographical districting,
Advances in Neural Information Processing Systems, 2024
Cheikh Ahmed , Alexandre Forel , Axel Parmentier , Thibaut Vidal
CF-OPT: Counterfactual explanations for structured prediction,
International Conference on Machine Learning, 2024
Germain Vivier-Ardisson , Alexandre Forel , Axel Parmentier , Thibaut Vidal
Don't explain noise: Robust counterfactuals for randomized ensembles,
International Conference on the Integration of Constraint Programming, Artificial Intelligence, and Operations Research, 2024
Alexandre Forel , Axel Parmentier , Thibaut Vidal
A survey of contextual optimization methods for decision making under uncertainty,
European Journal of Operational Research, 2024
Utsav Sadana , Abhilash Chenreddy , Erick Delage , Alexandre Forel , Emma Frejinger , Thibaut Vidal
Adaptive partitioning for chance-constrained problems with finite support,
Accepted at SIAM Journal on Optimization, 2023
Marius Roland , Alexandre Forel , Thibaut Vidal
Explainable data-driven optimization: From context to decision and back again,
International Conference on Machine Learning, 2023
Alexandre Forel , Axel Parmentier , Thibaut Vidal
Dynamic stochastic lot sizing with forecast evolution in rolling-horizon planning,
Production and Operations Management, 2023
Alexandre Forel , Martin Grunow

Preprints

The differentiable feasibility pump,
arXiv preprint arXiv:2411.03535, 2024
Matteo Cacciola , Alexandre Forel , Antonio Frangioni , Andrea Lodi
Free lunch in the forest: Functionally-identical pruning of boosted tree ensembles,
arXiv preprint arXiv:2408.16167, 2024
Youssouf Emine , Alexandre Forel , Idriss Malek , Thibaut Vidal