Home

About

Advanced Search

Browse by Discipline

Scientific Societies

E-print Alerts

Add E-prints

E-print Network
FAQHELPSITE MAPCONTACT US


  Advanced Search  

 
Learning the Structure of Deep Sparse Graphical Models Ryan Prescott Adams Hanna M. Wallach Zoubin Ghahramani
 

Summary: Learning the Structure of Deep Sparse Graphical Models
Ryan Prescott Adams Hanna M. Wallach Zoubin Ghahramani
University of Toronto University of Massachusetts Amherst University of Cambridge
Abstract
Deep belief networks are a powerful way
to model complex probability distributions.
However, it is di#cult to learn the structure
of a belief network, particularly one with hid­
den units. The Indian bu#et process has been
used as a nonparametric Bayesian prior on
the structure of a directed belief network with
a single infinitely wide hidden layer. Here, we
introduce the cascading Indian bu#et process
(CIBP), which provides a prior on the struc­
ture of a layered, directed belief network that
is unbounded in both depth and width, yet
allows tractable inference. We use the CIBP
prior with the nonlinear Gaussian belief net­
work framework to allow each unit to vary
its behavior between discrete and continuous

  

Source: Adams, Ryan Prescott - Department of Electrical and Computer Engineering, University of Toronto

 

Collections: Computer Technologies and Information Sciences