BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:From the Information Extraction Pipeline to Global Models\, and Ba
 ck - Sebastian Riedel\, UCL
DTSTART:20121016T130000Z
DTEND:20121016T140000Z
UID:TALK40987@talks.cam.ac.uk
CONTACT:Microsoft Research Cambridge Talks Admins
DESCRIPTION:Decisions in information extraction (IE)\, such as determining
  the types and relations of entities mentioned in text\, depend on each ot
 her. To remain efficient\, most systems make decisions in a sequential pip
 eline fashion\, even if later decisions could help earlier ones. In this t
 alk I will show how we used Conditional Random Fields to make these decisi
 ons jointly\, substantially outperformed less global approaches and ranked
  first in several international IE competitions. I will then present relax
 ation methods we developed and applied to scale up (exact) inference in su
 ch models. In the final part of my talk I will argue why we should not dis
 miss the pipeline and present an exact beam-search algorithm\, based on co
 lumn generation\, to overcome the pipeline's greedy nature.
LOCATION:Small public lecture room\, Microsoft Research Ltd\, 7 J J Thomso
 n Avenue (Off Madingley Road)\, Cambridge
END:VEVENT
END:VCALENDAR