BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//talks.cam.ac.uk//v3//EN
BEGIN:VTIMEZONE
TZID:Europe/London
BEGIN:DAYLIGHT
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
TZNAME:BST
DTSTART:19700329T010000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
TZNAME:GMT
DTSTART:19701025T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
CATEGORIES:NLIP Seminar Series
SUMMARY:Graph Neural Networks for Knowledge Base Question 
 Answering  - Daniil Sorokin\, Technische Universit
 ät Darmstadt
DTSTART;TZID=Europe/London:20181130T120000
DTEND;TZID=Europe/London:20181130T130000
UID:TALK114832AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/114832
DESCRIPTION:In this talk\, we present a semantic parsing appro
 ach to Knowledge Base Question Answering. We addre
 ss the problem of learning vector representations 
 for complex semantic parses that consist of multip
 le entities and relations. Previous work largely f
 ocused on selecting the correct semantic relations
  for a question and disregarded the structure of t
 he semantic parse: the connections between entitie
 s and the directions of the relations. We propose 
 to use Gated Graph Neural Networks to encode the g
 raph structure of the semantic parse. \n\nWe will 
 present a formulation of Gated Graph Neural Networ
 ks for labeled knowledge base subgraphs and show h
 ow it can be used in a question answering pipeline
 . Empirically\, we demonstrate on two data sets th
 at the graph networks outperform the baseline mode
 ls that do not explicitly model the semantic struc
 ture. 
LOCATION:FW11\, Computer Laboratory
CONTACT:Andrew Caines
END:VEVENT
END:VCALENDAR