BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Hierarchical Interpretation of Neural Text Classification - Prof. 
 Yulan He\, University of Warwick
DTSTART:20220224T110000Z
DTEND:20220224T120000Z
UID:TALK170564@talks.cam.ac.uk
CONTACT:Marinela Parovic
DESCRIPTION:Recent years have witnessed increasing interests in developing
  interpretable models in NLP. Most existing models aim at identifying inpu
 t features such as words or phrases important for model predictions. Neura
 l models developed in NLP however often compose word semantics in a hierar
 chical manner.  Interpretation by words or phrases only thus cannot faithf
 ully explain model decisions. In this talk\, I will present our recently p
 roposed Hierarchical Interpretable Neural Text classifier\, called Hint\, 
 which is able to identify the latent semantic factors and their compositio
 ns which contribute to the model's final decisions. This is often beyond w
 hat word-level interpretations could capture. Experimental results on both
  review datasets and news datasets show that our proposed approach achieve
 s text classification results on par with existing state-of-the-art text c
 lassifiers\, and generates interpretations more faithful to model predicti
 ons and better understood by humans than other interpretable neural text c
 lassifiers.
LOCATION:https://cam-ac-uk.zoom.us/j/97599459216?pwd=QTRsOWZCOXRTREVnbTJBd
 XVpOXFvdz09
END:VEVENT
END:VCALENDAR