2021

Heure:	10:30 - 11:30
Lieu:	Salle B107, bâtiment B, Université de Villetaneuse
Résumé:	Decomposition methods and column/matrix generation approaches for quadratic programming
Description:	Lucas Létocart The purpose of this talk is to present three decomposition approaches for quadratic problems. First, we analyze a simplicial decomposition like algorithmic framework that handles convex quadratic programs in an effective way. We also propose a branch & bound approach based on this simplicial decomposition for the binary case and we introduce warmstart techniques using columns' projection. Then, we propose a methodological analysis on a family of reformulations combining Dantzig-Wolfe decomposition and Quadratic Convex Reformulation principles for binary quadratic problems. Finally, we propose a matrix generation method for quadratically constrained 0-1 quadratic problems based on a Dantzig-Wolfe reformulation. The domain of this relaxation corresponds to the Boolean Quadric Polytope.