BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Cross-Lingual Word Embeddings in 60 Minutes - Ivan Vulić\, LTL\, 
 University of Cambridge
DTSTART:20181108T110000Z
DTEND:20181108T120000Z
UID:TALK114292@talks.cam.ac.uk
CONTACT:Edoardo Maria Ponti
DESCRIPTION:*Abstract*: In recent past\, NLP as a field has seen tremendou
 s utility of word embeddings as features in downstream tasks. The fact tha
 t these word vectors can be trained on unlabeled monolingual corpora of a 
 language makes them an inexpensive resource in NLP. With the increasing us
 e of monolingual word vectors\, there is a need for word vectors that can 
 be used as efficiently across multiple languages as monolingually. Therefo
 re\, learning bilingual and multilingual word embeddings is currently an i
 mportant research topic. These vectors offer an elegant and language-pair 
 independent way to represent content across different languages in shared 
 cross-lingual embedding spaces\, and also enable the integration of knowle
 dge from external resources (e.g.\, WordNet\, dictionaries) into the embed
 ding spaces. In this mini-tutorial\, I will briefly discuss the current te
 chniques in cross-lingual word embedding learning\, presenting the model t
 ypology based on multilingual training data requirements\, also including 
 very recent zero-supervision methods that require no bilingual data at all
 . I will then introduce several illustrative applications of the induced e
 mbedding spaces\, including bilingual dictionary induction\, ad-hoc cross-
 lingual information retrieval\, and cross-lingual transfer for dependency 
 parsing and dialogue state tracking.
LOCATION:Boardroom\, Faculty of English\, West Road
END:VEVENT
END:VCALENDAR