Home

About

Advanced Search

Browse by Discipline

Scientific Societies

E-print Alerts

Add E-prints

E-print Network
FAQHELPSITE MAPCONTACT US


  Advanced Search  

 
Slice sampling covariance hyperparameters of latent Gaussian models
 

Summary: Slice sampling covariance hyperparameters
of latent Gaussian models
Iain Murray
School of Informatics
University of Edinburgh
Ryan Prescott Adams
Dept. Computer Science
University of Toronto
Abstract
The Gaussian process (GP) is a popular way to specify dependencies be­
tween random variables in a probabilistic model. In the Bayesian framework
the covariance structure can be specified using unknown hyperparameters.
Integrating over these hyperparameters considers di#erent possible expla­
nations for the data when making predictions. This integration is often per­
formed using Markov chain Monte Carlo (MCMC) sampling. However, with
non­Gaussian observations standard hyperparameter sampling approaches
require careful tuning and may converge slowly. In this paper we present
a slice sampling approach that requires little tuning while mixing well in
both strong­ and weak­data regimes.
1 Introduction

  

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

 

Collections: Computer Technologies and Information Sciences