M7

M7:Regularization and Parameter Choice 

In this minisymposium regularization methods for the solution of ill-posed problems will be considered. Besides the classical regularization methods (Tikhonov method and its modifications, spectral cut-off, iteration methods etc) the actual topic is self-regularization where regularization is achieved by discretization of the ill-posed problem choosing proper discretization level as the regularization parameter depending on the noise level of the data. The choice of the regularization parameter is an important question in all regularization methods. The rules for the parameter choice are needed for different information of the noise level: there may be given the exact noise level or the rough estimate of it or no information about the noise level is given. Actual is theoretical and numerical comparison of accuracy of the regularized approximations to the solution.

Organizer:
U. Hämarik, University of Tartu, Estonia,   This email address is being protected from spambots. You need JavaScript enabled to view it.

Invited Speakers  

Uno Hämarik, University of Tartu, Estonia, This email address is being protected from spambots. You need JavaScript enabled to view it.
On comparison of accuracy of approximate solutions of operator equations with noisy data

Urve Kangro, University of Tartu, Estonia, This email address is being protected from spambots. You need JavaScript enabled to view it.
Solution of first kind cordial Volterra integral equations with noisy data

Marek Kojdecki, Warsaw Military University of Technology, Poland, This email address is being protected from spambots. You need JavaScript enabled to view it.
Algorithms for iterative a-posteriori choice of parameter in Tikhonov's regularisation

Robert Plato, University of Siegen, Germany, This email address is being protected from spambots. You need JavaScript enabled to view it.
The repeated midpoint rule for weakly singular Volterra equations of the first kind with noisy data

Toomas Raus, University of Tartu, Estonia, This email address is being protected from spambots. You need JavaScript enabled to view it.
Choice of the regularization parameter using local minimum points of the quasioptimality function

Kazimierz Reginski, Institute of Elektron Technology, Poland, This email address is being protected from spambots. You need JavaScript enabled to view it.
Regularization method for determining laser beam quality parameters

Carola-Bibiane Schönlieb, University of Cambridge, United Kingdom, This email address is being protected from spambots. You need JavaScript enabled to view it.
Bilevel optimisation for variational regularisation models