BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Structured prediction using energy-based models - Simon Lacoste-Ju
 lien (University of Cambridge)
DTSTART:20091008T130000Z
DTEND:20091008T143000Z
UID:TALK19983@talks.cam.ac.uk
CONTACT:Shakir Mohamed
DESCRIPTION:I will cover:\nYann LeCun\, Sumit Chopra\, Raia Hadsell\, Ranz
 ato Marc'Aurelio and Fu-Jie Huang: A Tutorial on Energy-Based Learning\, i
 n Bakir\, G. and Hofman\, T. and Schölkopf\, B. and Smola\, A. and Taskar
 \, B. (Eds)\, Predicting Structured Data\, MIT Press\, 2006\nhttp://yann.l
 ecun.com/exdb/publis/pdf/lecun-06.pdf\n\nLeCun's "Energy-Based Models" pro
 vide a unifying framework for several structured prediction approaches (wi
 th an intuitive motivation suitable to engineering new systems). I will us
 e this tutorial paper to introduce a few key ideas of structured predictio
 n. I will also go over a few aspects that LeCun is glossing over in his pr
 esentation (regularization e.g.).\n\nSection 8 is particularly interesting
  for food for thoughts on different approaches...\n\nMore info on energy-b
 ased models with slides\, videos\, etc. can be found at: http://www.cs.nyu
 .edu/~yann/research/ebm/
LOCATION:Engineering Department\, CBL Room 438
END:VEVENT
END:VCALENDAR