BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Transfer Learning for NLP - Sebastian Ruder\, INSIGHT Centre
DTSTART:20171013T130000Z
DTEND:20171013T140000Z
UID:TALK80741@talks.cam.ac.uk
CONTACT:Louise Segar
DESCRIPTION:Largely driven by Deep Learning\, over the course of the last 
 few years\, we have become more adept at training our models to map from i
 nputs to outputs with high precision. Our models break down\, however\, if
  the task or the data even slightly change and we are still at the beginni
 ng of learning how to transfer acquired knowledge. In this talk\, I will g
 ive an overview of transfer learning and look into its applications and pr
 omises for NLP. I will touch on multi-task learning as well as selecting r
 elevant data\, among other things."\n\nShort bio: Sebastian is a 2nd year 
 PhD Student in Natural Language Processing and Deep Learning at the Insigh
 t Research Centre for Data Analytics\, Dublin\, Ireland and a research sci
 entist at Dublin-based text analytics startup AYLIEN. He previously studie
 d Computational Linguistics at the University of Heidelberg\, Germany and 
 at Trinity College\, Dublin. During his studies\, he's worked with Microso
 ft\, IBM's Extreme Blue\, Google Summer of Code\, and SAP\, among others.
LOCATION:JDB Seminar Room\, CUED
END:VEVENT
END:VCALENDAR