Deep Gaussian Processes
Add to your list(s)
Download to your calendar using vCal
If you have a question about this talk, please contact Zoubin Ghahramani.
[NOTE: rescheduled, now 15:00-16:00]
In this talk we will introduce deep Gaussian process (GP) models. Deep
GPs are a deep belief network based on Gaussian process mappings. The
data is modeled as the output of a multivariate GP. The inputs to that
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 model which we use for model selection (number of
layers and nodes per layer). Deep belief networks are typically
applied to relatively large data sets using stochastic gradient
descent for optimization. Our fully Bayesian treatment allows for the
application of deep models even when data is scarce. Model selection
by our variational bound shows that a five layer hierarchy is
justified even when modelling a digit data set containing only 150
examples. In the seminar we will briefly review dimensionality reduction
via Gaussian processes, before showing how this framework can be
extended to build deep models.
This talk is part of the Machine Learning @ CUED series.
This talk is included in these lists:
Note that ex-directory lists are not shown.
|