BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Multilingual Models for Distributed Semantics - Karl Moritz Herman
 n\, Oxford University
DTSTART:20140606T110000Z
DTEND:20140606T120000Z
UID:TALK52237@talks.cam.ac.uk
CONTACT:Tamara Polajnar
DESCRIPTION:In this talk I will present a technique for learning semantic 
 representations\, which extends the distributional hypothesis to multiling
 ual data and joint-space embeddings. These models leverage parallel data a
 nd learn to strongly align the embeddings of semantically equivalent sente
 nces\, while maintaining sufficient distance between those of dissimilar s
 entences\, using a form of noise-contrastive update.\n\nA nice feature of 
 these models is that they do not rely on word alignments or any syntactic 
 information\, making them easy to apply to a large number of diverse langu
 ages. I will briefly also describe an extension of this approach to learn 
 semantic representations at the document level. \n\nThe talk will conclude
  with an analysis of these models and some empirical evaluation. Using sev
 eral cross-lingual document classification tasks\, I show that this approa
 ch can be used to learn semantically plausible\, multilingual distributed 
 representations.
LOCATION:FW26\, Computer Laboratory
END:VEVENT
END:VCALENDAR