BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4//
BEGIN:VEVENT
UID:20251104T023733EST-8630PhNbML@132.216.98.100
DTSTAMP:20251104T073733Z
DESCRIPTION:Generalized Sparse Additive Models\n\nI will present a unified 
 approach to the estimation of generalized sparse additive models in high d
 imensional regression problems. Our approach is based on combining structu
 re-inducing and sparsity penalties in a single regression problem. It allo
 ws for the use of a large family of structure-inducing penalties: Those ch
 aracterized by semi-norm constraints. This includes finite dimensional lin
 ear subspaces\, sobolev and holder classes\, classes with bounded total va
 riation\, among others. We give an efficient computational algorithm to fi
 t this family of models that easily scales to thousands of observations an
 d features. In addition we develop a framework for proving convergence bou
 nds on these estimators\; and show that our estimators converge at the min
 imax optimal rate under suitable conditions. We also compare the performan
 ce of existing methods in an empirical study and discuss directions for fu
 ture work.\n\n \n\n \n\n \n
DTSTART:20180119T203000Z
DTEND:20180119T213000Z
LOCATION:Room 1205\, Burnside Hall\, CA\, QC\, Montreal\, H3A 0B9\, 805 rue
  Sherbrooke Ouest
SUMMARY:Asad Haris PhD Candidate Department of Biostatistics University of 
 Washington
URL:/mathstat/channels/event/asad-haris-phd-candidate-
 department-biostatistics-university-washington-283865
END:VEVENT
END:VCALENDAR