lundi 2 juin 14h00 Intervenant: Lothar Reichel (Kent State university, USA) ----------- Titre: « Matrix Decompositions for Tikhonov Regularization » ------ Résumé: ------- We discuss several generalized Krylov subspace methods for reducing a pair of large matrices {A,B} to a pair of small matrices {H,K}. Generalizations of Golub-Kahan bidiagonalization and the Arnoldi process are considered. Applications to Tikhonov regularization of large discrete ill-posed problems are described, in which A is a discretization of a compact operator and B a discretized differential operator.