BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Deep Gaussian Processes - Neil Lawrence\, University of Sheffield
DTSTART:20130903T140000Z
DTEND:20130903T150000Z
UID:TALK46963@talks.cam.ac.uk
CONTACT:Microsoft Research Cambridge Talks Admins
DESCRIPTION:In this talk we will introduce deep Gaussian process (GP) mode
 ls. Deep GPs are a deep belief network based on Gaussian process mappings.
  The data is modelled as the output of a multivariate GP. The inputs to th
 at Gaussian process are then governed by another GP. A single layer model 
 is equivalent to a standard GP or the GP latent variable model(GPLVM). We 
 perform inference in the model by approximate variational marginalization.
  This results in a strict lower bound on the marginal likelihood of the mo
 del which we use for model selection (number of layers and nodes per layer
 ). Deep belief networks are typically applied to relatively large data set
 s using stochastic gradient descent for optimization. Our fully Bayesian t
 reatment allows for the application of deep models even when data is scarc
 e. Model selection by our variational bound shows that a five layer hierar
 chy is justified even when modelling a digit data set containing only 150e
 xamples. In the seminar we will briefly review dimensionality reduction vi
 a Gaussian processes\, before showing how this framework can be extended t
 o build deep models.
LOCATION:Auditorium\, Microsoft Research Ltd\, 21 Station Road\, Cambridge
 \, CB1 2FB
END:VEVENT
END:VCALENDAR