First presenter			Co-presenter(s)
Name :	Benoit Lacelle *		Name:	Eric Schost *
E-mail:			E-mail:
Affiliation:	University of Western Ontario		Name:
Department:	Computer Science Department		E-mail:
City:			Name:
State/Province:			E-mail:
Country:	Canada		Name:
Talk Number:	11-02		E-mail:
Session:	11- High-Performance Computer Algebra		Schedule:   Room:	Thursday, 16:30   B-2624
Related website:
Title of presentation:	Towards an efficient implementation for the resolution of structured linear system
Abstract:
Lots of linear algebra problems can be reduced to the resolution of a linear system: A.X = B. When they are expressed in such a form, it appears that A often admits a pattern : it is said to be a structured matrix. To accelerate its resolution, one can take advantage of that structure. This talk will present our efficient implementation of the Morf-Bitmead-Anderson algorithm for the inversion of scalar structured matrices : its time complexity is quasi-linear in the size of A. That implementation has been associated with a Newton Iteration taking advantage of the structure of A : it is able to inverse polynomial structured matrices with again a time complexity quasi-linear in the size of A.