BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Handling Sparsity via the Horseshoe - Carlos Carvalho\, University
  of Chicago Booth School of Business
DTSTART:20090424T150000Z
DTEND:20090424T160000Z
UID:TALK18114@talks.cam.ac.uk
CONTACT:8047
DESCRIPTION: This paper presents a general\, fully Bayesian framework for\
 nsparse supervised-learning problems based on the horseshoe prior. The\nho
 rseshoe prior is a member of the family of multivariate scale\nmixtures of
  normals\, and is therefore closely related to widely used\napproaches for
  sparse Bayesian learning\, including\, among others\,\nLaplacian (LASSO) 
 and Student-t priors (relevance vector machines).\nThe advantages of the\n
 horseshoe are its robustness at handling unknown sparsity and large\noutly
 ing signals. These properties are justified theoretically via a\nrepresent
 ation theorem and accompanied by comprehensive empirical\nexperiments that
  compare its performance to benchmark alternatives.\n
LOCATION:MR12\, CMS\, Wilberforce Road\, Cambridge\, CB3 0WB
END:VEVENT
END:VCALENDAR