BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Sparse Approximation and Atomic Decomposition: Considering Atom In
 teractions in Evaluating and Building Signal Representations - Bob Sturm\,
  Chateaubriand Fellow post-doctoral researcher at UPMC - Paris 06 with Pro
 fessor Laurent Daudet
DTSTART:20090702T103000Z
DTEND:20090702T113000Z
UID:TALK18789@talks.cam.ac.uk
CONTACT:Rachel Fogg
DESCRIPTION:I will present work from my recent dissertation\, which makes 
 contributions to the sparse approximation and efficient representation of 
 complex signals\, e.g.\, acoustic signals\, using greedy iterative descent
  pursuits and overcomplete dictionaries. As others have noted before\, pec
 uliar problems arise when a signal model is mismatched to the signal conte
 nt\, and a pursuit makes bad selections from the dictionary. These result 
 in a model that contains atoms having no physical significance to the sign
 al\, existing to correct the representation through destructive interferen
 ce\, which diminishes its efficiency\, and hinder the useful application o
 f sparse approximation to signal analysis (e.g.\, source identification)\,
  visualization (e.g.\, source selection)\, and modification (e.g.\, source
  extraction). While past works have addressed these problems by reformulat
 ing a pursuit to avoid them\, in this dissertation we use these corrective
  terms to learn about the signal\, the pursuit algorithm\, the dictionary\
 , and the model. We show that a better signal model results when a pursuit
  builds it considering the interactions between its components.\n\n
LOCATION:LR11\, Engineering\, Department of
END:VEVENT
END:VCALENDAR