Home

About

Advanced Search

Browse by Discipline

Scientific Societies

E-print Alerts

Add E-prints

E-print Network
FAQHELPSITE MAPCONTACT US


  Advanced Search  

 
Gaussian Process Product Models for Nonparametric Nonstationarity
 

Summary: Gaussian Process Product Models for Nonparametric
Nonstationarity
Ryan Prescott Adams rpa23@cam.ac.uk
Oliver Stegle os252@cam.ac.uk
Cavendish Laboratory, University of Cambridge, Cambridge CB3 0HE, UK
Abstract
Stationarity is often an unrealistic prior as-
sumption for Gaussian process regression.
One solution is to predefine an explicit non-
stationary covariance function, but such co-
variance functions can be difficult to spec-
ify and require detailed prior knowledge of
the nonstationarity. We propose the Gaus-
sian process product model (GPPM) which
models data as the pointwise product of two
latent Gaussian processes to nonparametri-
cally infer nonstationary variations of ampli-
tude. This approach differs from other non-
parametric approaches to covariance function
inference in that it operates on the outputs

  

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

 

Collections: Computer Technologies and Information Sciences