BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Human Behavior Classification with Infinite Hidden Conditional Ran
 dom Fields - Konstantinos Bousmalis and Stefanos Zafeiriou (Imperial Colle
 ge)
DTSTART:20120727T100000Z
DTEND:20120727T110000Z
UID:TALK39129@talks.cam.ac.uk
CONTACT:Zoubin Ghahramani
DESCRIPTION:Hidden Conditional Random Fields (HCRFs) are discriminative la
 tent variable models which have been shown to successfully learn the hidde
 n structure of a given classification problem (provided an appropriate val
 idation of the number of hidden states). The Infinite Hidden Conditional R
 andom Field (IHCRF) is a Hierarchical Dirichlet Process-Hidden Conditional
  Random Field with a countably infinite number of hidden states\, which ri
 ds us not only of the necessity to specify a priori a fixed number of hidd
 en states available to the latent variables of the model but also of the p
 roblem of overfitting. In this talk\, we will present the model and two ap
 proaches to learning it: an effective Markov chain Monte Carlo (MCMC) samp
 ling technique\, and a novel variational approach. We show that the iHCRF 
 is able to converge to a correct number of represented hidden states\, and
  outperforms the best finite HCRFs -chosen via cross-validation- for the d
 ifficult tasks of recognizing instances of spontaneous agreement\, disagre
 ement\, and pain. We assume the audience has a basic understanding of Diri
 chlet Processes.
LOCATION:Engineering Department\, CBL Room BE-438
END:VEVENT
END:VCALENDAR