Header logo is

Approximate Inference for Robust Gaussian Process Regression

2005

Technical Report

ei


Gaussian process (GP) priors have been successfully used in non-parametric Bayesian regression and classification models. Inference can be performed analytically only for the regression model with Gaussian noise. For all other likelihood models inference is intractable and various approximation techniques have been proposed. In recent years expectation-propagation (EP) has been developed as a general method for approximate inference. This article provides a general summary of how expectation-propagation can be used for approximate inference in Gaussian process models. Furthermore we present a case study describing its implementation for a new robust variant of Gaussian process regression. To gain further insights into the quality of the EP approximation we present experiments in which we compare to results obtained by Markov chain Monte Carlo (MCMC) sampling.

Author(s): Kuss, M. and Pfingsten, T. and Csato, L. and Rasmussen, CE.
Number (issue): 136
Year: 2005
Day: 0

Department(s): Empirical Inference
Bibtex Type: Technical Report (techreport)

Institution: Max Planck Institute for Biological Cybernetics, Tübingen, Germany

Organization: Max-Planck-Gesellschaft
School: Biologische Kybernetik

Links: PDF

BibTex

@techreport{3265,
  title = {Approximate Inference for Robust Gaussian Process Regression},
  author = {Kuss, M. and Pfingsten, T. and Csato, L. and Rasmussen, CE.},
  number = {136},
  organization = {Max-Planck-Gesellschaft},
  institution = {Max Planck Institute for Biological Cybernetics, T{\"u}bingen, Germany},
  school = {Biologische Kybernetik},
  year = {2005},
  doi = {}
}