The department of Mathematics and Statistics at the University of Limerick invites you to a seminar by Dr Kirk Soodhalter (Trinity College Dublin):
Title: Krylov Subspace Methods for large-scale computing and imagine reconstruction
Abstract:. Krylov subspace methods are a powerful class of iterative methods for sparse linear problems in which the operator may be available only through a procedure implementing its application. In this talk, we review the core mathematics underpinning the entire class of methods before pivoting to discuss some of our recent work in this area, focusing on the analysis of Block Krylov Subspace methods, augmented methods (a class of methods allowing one to use a priori known data to drive the iteration, and methods for large-scale ill-posed problems. For the latter, we focus on solving ill-posed image reconstruction problems for which the exact convolution operator is not know (i.e., a blind deconvolution problem). Numerical experiments will be used to demonstrate the potential for these methods in various applications.
This seminar will take place on Friday 9th March , at 4 p.m, in Room A2-002.
If you have any questions regarding this seminar, please direct them to Iain Moyles (061 233726, email@example.com).
A full list of upcoming seminars can be found at http://www.ulsites.ul.ie/macsi/node/48011
Supported by Science Foundation Ireland funding, MACSI - the Mathematics Applications Consortium for Science and Industry (www.macsi.ul.ie), centred at the University of Limerick, is dedicated to the mathematical modelling and solution of problems which arise in science, engineering and industry in Ireland.