PhD AWARD  2022

 

  • Adèle Pass-Lanneau

Adèle Pass-Lanneau completed her PhD at Sorbonne University, under the supervision of Pascale Bendotti, Philippe Chrétienne and Pierre Fouilhoux. Her thesis work deals with the notion of decision anchoring in the field of robust combinatorial optimization. She proposed a new criterion, called anchoring, to favor unchanged decisions between solutions, in reoptimization and robust optimization. The proposed robust-anchored approach allows to find a trade-off between the cost of the solution and the guarantees on the decisions under uncertainty.
The anchoring problems are formalized and declined on two classes of problems: linear problems in binary variables, and project scheduling problems under precedence or resource constraints. An industrial application of this second class is maintenance scheduling at EDF. The algorithmic complexity of anchoring problems is analyzed. The fine-grained study of the combinatorial properties of anchored solutions allows to design dedicated and efficient algorithmic and polyhedral approaches. Integer linear programming techniques are implemented, demonstrating the implementability of anchoring problems.
Adèle Pass-Lanneau's thesis work is remarkable from both a mathematical and a computational point of view. It has already led to several scientific publications in internationally renowned journals.  
Adèle Pass-Lanneau has shown great scientific maturity in the treatment of the problems studied and the thesis demonstrates a strong intuition and an exceptional mastery of the tools of theory and practice in mathematics, computer science and operations research.

  • Étienne Boursier

Etienne Boursier completed his PhD in mathematics at ENS Paris Saclay under the supervision of Vianney Perchet. He worked on different learning problems when certain interactions between intelligent agents occur. In particular, he studied how sequential learning algorithms can behave in strategic environments when confronted with other agents. He proposed computationally efficient algorithms with good performance guarantees (low regret) for various problems: the multi-player bandit problem is the main focus of his thesis. Queuing systems, online repeated auctions and social learning have also been studied.
Etienne Boursier has written a very rich and excellent thesis. His work makes important contributions to sequential learning. The numerous results obtained cover a wide and coherent spectrum of topics that required the development of original methods combining probabilistic tools and techniques from game theory and optimization.

 

The jury for the 2022 edition was chaired by Samir Adly, Université de Limoges and composed of :

Members appointed by the PGMO Scientific Council

Samuel Vaiter (Université Côte d'Azur)
Guillaume Vigeral (Université Paris Dauphine)
Edouard Pauwels (Université Toulouse 3 Paul Sabatier)

Members appointed by the ROADEF

Ruslan Sadykov (INRIA)
Laetitia Jourdan (Université de Lille)
Bruno Escoffier (Sorbonne Université)

Members appointed by the group MODE of SMAI

Samir Adly (Université de Limoges)
Charles Dossal (INSA)
Chloé Jimenez (Université de Brest)