BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Unsupervised Learning with Latent Variable Models - Zhenwen Dai
DTSTART:20150616T100000Z
DTEND:20150616T110000Z
UID:TALK59905@talks.cam.ac.uk
CONTACT:12852
DESCRIPTION:Unsupervised learning is under rapid development. The probabil
 istic approach is typically based on latent variable models. In this prese
 ntation\, I will show a number of latent variable models that I have worke
 d on\, spaning from parametric to non-parametric and from linear to non-li
 near. I will show the connection between these models and link to my most 
 recent work: infinite dimensional Gaussian process latent variable models 
 and variational hierarchical communities of experts. I will derive variati
 onal lower bounds of these models for efficient inference and show some ap
 plications of the developed models on real data.
LOCATION:Engineering Department\, CBL Room BE-438
END:VEVENT
END:VCALENDAR