BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Approximate Bayesian Inference : Relaxations\, Algorithms and Larg
 e Scale Applications
DTSTART:20100628T160000
DTSTAMP:20260407T210755Z
UID:81ae14b0be8de2a46bc894f80639123bd5b24418bd8abb1c4e16daf3
CATEGORIES:Conferences - Seminars
DESCRIPTION:Matthias Seeger\, Saarland University and Max Planck Institute
  for Informatics\nAbstract:\n\nToday's real world problems of large scale 
 information processing demand\ndecision making from uncertain knowledge\, 
 not only in high level domains like\nlanguage understanding or intelligent
  behaviour\, but increasingly so in\nfundamental fields of signal and imag
 e processing\, if only to circumvent\nspiralling hardware costs or statist
 ical limits coming with classical\napproaches. Modern science\nand medicin
 e\, with questions growing more rapidly than data streams\, need\nautonomo
 us tools for acquisition and sifting of data or experimental planning.\nBa
 yesian graphical modelling is the preeminent language and calculus for\nwo
 rking with uncertain information\, but poses hard computational challenges
  in\npractice. Some of these have successfully been addressed in machine l
 earning\,\ncalling on ideas from convex optimization\, numerical mathemati
 cs\, and graph theory.\nMore information on : http://people.mmci.uni-saarl
 and.de/~mseeger/
LOCATION:INM200
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR