Iterative methods for Tikhonov regularization

Lothar Reichel
Kent State University
Mathematics

The solution of large-scale ill-posed problems has recently received considerable attention. Many of the available solution methods are based on Tikhonov regularization. This talk surveys iterative methods for Tikhonov regularization that explore the connection between orthogonal polynomials, Gauss quadrature, and Lanczos bidiagonalization. Methods for unconstrained and
constrained regularization problems will be discussed and comparisons with other available iterative methods will be presented. The talk presents joint work with Daniela Calvetti, Gene Golub, Per Christian Hansen, Bryan Lewis,
and Fiorella Sgallari.


Back to Long Programs