BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Structured Learning and Structural SVMs - S-X. Austin Zhang and Ju
 an Pino
DTSTART:20100610T130000Z
DTEND:20100610T143000Z
UID:TALK24006@talks.cam.ac.uk
CONTACT:Shakir Mohamed
DESCRIPTION:This talk is an introduction to structured learning\, more spe
 cifically we focus on Structural SVMs (SSVMs). We introduce the concept of
  structure\, then we present the SSVMS formulation\, training algorithms a
 nd connection to other existing models. We then present two applications: 
 parsing and word alignment.\n\nIf you don't have time\, read the last thre
 e papers and only skim through the first paper ("Cutting Plane\, etc."). I
  you have time to read only one paper\, you can read "Support Vector Machi
 ne Learning for Interdependent and Structured Output Spaces".\n\n"Paper 1"
 :http://mi.eng.cam.ac.uk/~jmp84/temp/RCC/Cutting-Plane%20Training%20of%20S
 tructural%20SVMs_JML%2009_structSVM.pdf\n\n"Paper 2":http://mi.eng.cam.ac.
 uk/~jmp84/temp/RCC/Learning%20Structural%20SVMs%20with%20Latent%20Variable
 s.pdf\n\n"Paper 3":http://mi.eng.cam.ac.uk/~jmp84/temp/RCC/Learning%20stru
 ctured%20prediction%20Models__A%20large%20margin%20approach.pdf\n\n"Paper 
 4":http://mi.eng.cam.ac.uk/~jmp84/temp/RCC/Support%20Vector%20Learning%20f
 or%20Interdependent%20and%20Structured%20Output%20Spaces__ICML%202004.pdf\
 n\n\n
LOCATION:Engineering Department\, CBL Room 438
END:VEVENT
END:VCALENDAR