BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Learning to Generate Textual Data - Pontus Stenetorp (UCL)
DTSTART:20170126T110000Z
DTEND:20170126T120000Z
UID:TALK69541@talks.cam.ac.uk
CONTACT:Mohammad Taher Pilehvar
DESCRIPTION:To learn text understanding models with millions of parameters
  one needs massive amounts of data.  In this work\, we argue that generati
 ng data can compensate for this need.  While defining generic data generat
 ors is difficult\, we propose to allow generators to be "weakly" specified
  in the sense that a set of parameters controls how the data is generated.
   Consider for example generators where the example templates\, grammar\, 
 and/or vocabulary is determined by this set of parameters.  Instead of man
 ually tuning these parameters\, we learn them from the limited training da
 ta at our disposal.  To achieve this\, we derive an efficient algorithm ca
 lled GeneRe that jointly estimates the parameters of the model and the und
 etermined generation parameters.  We illustrate its benefits by learning t
 o solve math exam questions using a highly parametrised sequence-to-sequen
 ce neural network.
LOCATION: SR-24\, English Faculty Building\, 9 West Road (Sidgwick Site)
END:VEVENT
END:VCALENDAR