Title : Métaheuristique basée sur le recuit quantique

Speakers : Elèves de deuxième année, cycle-ingénieur Supélec 
           (MM. Belaube, de Méric de Bellefon, Bergé, Cavarec, 
           Chabrillac, Couronné, Doutre, Dufoulon, Lundy, 
           Salaün)

 

Abstract:

We have been working on a scientific project during an entire
 academic year. Our goal was to understand and, possibly, 
improve a new and yet not well known metaheuristic 
optimization method, quantum annealing.

This algorithm is inspired by simulated annealing and reviewed
 from the quantum physics perspective. It deals with difficult
 problems more efficiently than the classical annealing.

We explain the algorithm principle and discuss the results 
obtained for the reference problems (vertex coloring, TSP). 
We present our adaptations of this algorithm to solve the 
k-center problem and the portfolio optimization problem. 
We show that it is possible to improve the quantum annealing 
method by reducing the search space dynamically.