Difference between revisions of "Rodolphe Sepulchre, June 2013"

From MurrayWiki
Jump to: navigation, search
(Abstract)
(Agenda)
Line 5: Line 5:
 
{{agenda item|9:30a|Richard Murray, 109 Steele Lab}}
 
{{agenda item|9:30a|Richard Murray, 109 Steele Lab}}
 
{{agenda item|9:45a|Meet with Richard's NCS group, 110 Steele}}
 
{{agenda item|9:45a|Meet with Richard's NCS group, 110 Steele}}
* 9:45-10:45: Necmiye, Mumu, Eric, Ivan
+
* 9:45-10:45: Necmiye, Mumu, Eric, Rangoli
 
* 10:45-11:45: Enoch, Marcella, Anandh, Dan
 
* 10:45-11:45: Enoch, Marcella, Anandh, Dan
 
{{agenda item|11:45|Seminar setup}}
 
{{agenda item|11:45|Seminar setup}}

Revision as of 04:14, 30 May 2013

Rodolphe Sepulchre will visit Caltech on 3 June 2013 (Mon).

Agenda

9:30a   Richard Murray, 109 Steele Lab
9:45a   Meet with Richard's NCS group, 110 Steele
  • 9:45-10:45: Necmiye, Mumu, Eric, Rangoli
  • 10:45-11:45: Enoch, Marcella, Anandh, Dan
11:45   Seminar setup
12:00p   Seminar, 213 ANB
1:00p   Lunch with Venkat, CMS faculty
2:15p   Venkat Chandrasekaran, 300 Annenberg
3:00p   Open
3:45p   Open
4:30p   Andrea Censi)
5:15p   Done

Abstract

The geometry of (thin) SVD revisited for large-scale computations

Rodolphe Sepulchre
University of Liege, Belgium

The talk will introduce a riemannian framework for large-scale computations over the set of low-rank matrices. The foundation is geometric and the motivation is algorithmic, with a bias towards efficient computations in large-scale problems. We will explore how classical matrix factorizations connect the riemannian geometry of the set of fixed-rank matrices to two well-studied manifolds: the Grassmann manifold of linear subspaces and the cone of positive definite matrices. The theory will be illustrated on various applications, including low-rank Kalman filtering, linear regression with low-rank priors, matrix completion, and the choice of a suitable metric for Diffusion Tensor Imaging.