The ORKAD team aims to exploit simultaneously expertise in combinatorial optimization and knowledge extraction to address upcoming optimization problems. While the two scientific areas are developed more or less independently, the synergy between combinatorial optimization and knowledge extraction offers an opportunity first, to improve the performance and autonomy of optimization methods thanks to Knowledge and secondly to solve efficiently Knowledge extraction problems thanks to operational research methods. Our approaches are mainly based on mono and multi-objective combinatorial optimization and led to the development of open source software.
Laetitia Jourdan
Sélection et configuration d'algorithmes pour la construction d'emplois du temps universitaires
Nouveaux algorithmes de prédiction et de planification pour le digital learning basés sur des méthodes d'optimisation
MO-AutoML : un framework multiobjectif pour la configuration automatique d'algorithmes de machine learning
Automatic Design of Dynamic Local Search Algorithms
Configuration automatique de réseaux de neurones profonds, à l'aide de méthodes multi-objectifs
Cadre général et méthodes d'optimisation pour la planification de stock en contextes industriels 27/08/2020
Optimizing competitive economic decisions in a business game 13/10/2017