University of Leicester
Browse
multi-output-gp-v4.pdf (188.72 kB)

Gaussian process regression with multiple response variables

Download (188.72 kB)
journal contribution
posted on 2015-03-04, 15:57 authored by Bo Wang, Tau Chen
Gaussian process regression (GPR) is a Bayesian non-parametric technology that has gained extensive application in data-based modelling of various systems, including those of interest to chemometrics. However, most GPR implementations model only a single response variable, due to the difficulty in the formulation of covariance function for correlated multiple response variables, which describes not only the correlation between data points, but also the correlation between responses. In the paper we propose a direct formulation of the covariance function for multi-response GPR, based on the idea that its covariance function is assumed to be the “nominal” uni-output covariance multiplied by the covariances between different outputs. The effectiveness of the proposed multi-response GPR method is illustrated through numerical examples and response surface modelling of a catalytic reaction process.

History

Citation

Chemometrics and Intelligent Laboratory Systems 142 (2015) 159–165

Author affiliation

/Organisation/COLLEGE OF SCIENCE AND ENGINEERING/Department of Mathematics

Version

  • AM (Accepted Manuscript)

Published in

Chemometrics and Intelligent Laboratory Systems 142 (2015) 159–165

Publisher

Elsevier for Chemometrics Society

issn

0169-7439

Copyright date

2015

Available date

2017-02-02

Publisher version

http://www.sciencedirect.com/science/article/pii/S0169743915000180

Language

en

Usage metrics

    University of Leicester Publications

    Categories

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC