BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Building a True Semantic World: Generalizing set-theoretic semanti
 cs in vector spaces - Eva Maria Vecchi
DTSTART:20170511T100000Z
DTEND:20170511T110000Z
UID:TALK71803@talks.cam.ac.uk
CONTACT:Mohammad Taher Pilehvar
DESCRIPTION:Quantification is intrinsic to most utterances in natural lang
 uage. It encompasses fundamental information which is a prerequisite for l
 exical semantics and inference tasks\, such as hyponomy and entailment. Ye
 t\, we are unable to denote this information computationally in state-of-t
 he-art semantic models. In this talk\, I present an approach to automatica
 lly map a distributional semantic\nspace onto a set-theoretic model to ind
 uce generalized quantifiers for subject-predicate pairs. I begin from the 
 prediction that there is a functional relationship between distributional 
 information and vectorial concept representations. I first introduce a lar
 ge\, computationally-friendly dataset motivated by understanding\, theoret
 ically\, to which extent speakers agree on a single model of the world. Th
 en I present a model to test this prediction and show that we can\, indeed
 \, map between formalisms\, and further\, we can generate natural language
  quantifiers sensibly.
LOCATION: SR-24\, English Faculty Building\, 9 West Road (Sidgwick Site)
END:VEVENT
END:VCALENDAR