BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Graph Neural Networks for Knowledge Base Question Answering  - Dan
 iil Sorokin\, Technische Universität Darmstadt
DTSTART:20181130T120000Z
DTEND:20181130T130000Z
UID:TALK114832@talks.cam.ac.uk
CONTACT:Andrew Caines
DESCRIPTION:In this talk\, we present a semantic parsing approach to Knowl
 edge Base Question Answering. We address the problem of learning vector re
 presentations for complex semantic parses that consist of multiple entitie
 s and relations. Previous work largely focused on selecting the correct se
 mantic relations for a question and disregarded the structure of the seman
 tic parse: the connections between entities and the directions of the rela
 tions. We propose to use Gated Graph Neural Networks to encode the graph s
 tructure of the semantic parse. \n\nWe will present a formulation of Gated
  Graph Neural Networks for labeled knowledge base subgraphs and show how i
 t can be used in a question answering pipeline. Empirically\, we demonstra
 te on two data sets that the graph networks outperform the baseline models
  that do not explicitly model the semantic structure. 
LOCATION:FW11\, Computer Laboratory
END:VEVENT
END:VCALENDAR