Research

Machine Learning for Optimization

My research focuses on the intersection of Operations Research and Machine Learning. I already work on specialized Machine Learning techniques within the context of Lagrangian Relaxation and the unrolling of the Bundle method.

Machine Learning for the Lagrangian Relaxation

Lagrangian relaxation stands among the most efficient approaches for solving Mixed Integer Linear Programs (MILPs) with difficult constraints. Given any duals for these constraints, called Lagrangian Multipliers (LMs), it returns a bound on the optimal value of the MILP, and Lagrangian methods seek the LMs giving the best such bound. But these methods generally rely on iterative algorithms resembling gradient descent to maximize the concave piecewise linear dual function: the computational burden grows quickly with the number of relaxed constraints.

Unrolling Techniques

The concept of Algorithm Unrolling entails the transformation of an iterative algorithm's execution into a continuous differential representation. This adaptation facilitates the incorporation of neural networks within the algorithm's execution, while retaining the ability to compute gradients for subsequent backpropagation steps. Currently, my research focuses on refining this technique within the framework of the Bundle Algorithm, a highly efficient iterative method commonly employed for optimizing piecewise linear functions.

Publications

Predicting Lagrangian Multipliers for Mixed Integer Linear Programs

Francesco Demelas, Joseph Le Roux, Mathieu Lacroix, Axel Parmentier (2024)
International Conference of Machine Learning, (ICML)
Vienna, Austria, July 21-27.

more

Teaching

Institut Galilee

2021-2022
  • HTML 5 et CSS 3 1ère année
  • Introduction à l'Environment Unix 1ère année
  • Programmation 1 1ère année
2022-2023
  • HTML 5 et CSS 3 1ère année
  • Introduction à l'Environment Unix 1ère année
  • Programmation 1 1ère année
2023/2024
  • HTML 5 et CSS 3 1ère année
  • Remise à niveaux en informatique
  • Structures de données et algorithmes 2ème année,