Henri Lefebvre

I am a CNRS researcher working on mathematical optimization at the LIRMM in Montpellier (France). I received my PhD in Operations Research from the University of Bologna in 2023, under the supervision of Michele Monaci and Enrico Malaguti. My doctoral research focused on optimization problems under uncertainty, with a particular emphasis on robust optimization. In 2024, I was awarded the EURO Best Dissertation Award by the Association of European Operational Research Societies. I then worked as a postdoctoral researcher at the University of Trier (Germany) in the working group of Martin Schmidt.

My research focuses on the development of algorithms for complex decision-making problems involving uncertainty and multiple decision-makers. I aim to design generic approaches applicable to broad classes of problems, which can be adapted to specific real-world scenarios.

Research Interests

Connected research areas Four overlapping fields of research Optimization under uncertainty, explainability, bilevel optimization, and game theory. Research topics below explore these fields and their intersections. Optimizationunder uncertainty Bileveloptimization Game theory Explainability

7 topics across four connected fields

Optimization under Uncertainty

  • Adjustable robust optimization
  • Decomposition methods
  • Information discovery

Bilevel Optimization

  • Single-level reformulations
  • Cutting planes and heuristics
  • Calmness and coupling constraints

Uncertainty ∩ Bilevel

  • Decision-dependent uncertainty
  • Wait-and-see followers
  • Column-and-constraint generation

Uncertainty ∩ Explainability ∩ Bilevel

  • Two-stage stochastic programs
  • First-stage decision explanations
  • Scenario reduction

Uncertainty ∩ Game Theory

  • Robust games
  • Extensive-form models

Explainability ∩ Bilevel

  • Counterfactual explanations
  • Bilevel models
  • Penalty alternating direction methods

Game Theory

  • Generalized Nash equilibria
  • Equilibrium reductions
  • Variational inequalities
Research software

idol A C++ framework for optimization

Build custom algorithms for mixed-integer, robust, and bilevel optimization by combining decomposition methods.

Compose your algorithmC++
const auto branch_and_price =
    branch_and_bound + column_generation;

model.use(branch_and_price);
model.optimize();

Education

Doctorate

PhD in Operations Research

Università di Bologna DEI · Italy
Doctoral thesis

Adjustable Robust Optimization with Nonlinear Recourses

Supervisors: Michele Monaci and Enrico Malaguti

Earlier degrees 2019 · France & Italy
Engineering & science
Université de Technologie de Compiègne France

Master of Engineering

Decision support in logistics

Minor: Philosophy, technology and cognition

Master of Science

Optimization and learning of systems