Analysis of some methods for reduced rank Gaussian process regression

J. Quinonero-Candela, Carl Edward Rasmussen

    Research output: Chapter in Book/Report/Conference proceedingBook chapterResearchpeer-review


    While there is strong motivation for using Gaussian Processes (GPs) due to their excellent performance in regression and classification problems, their computational complexity makes them impractical when the size of the training set exceeds a few thousand cases. This has motivated the recent proliferation of a number of cost-effective approximations to GPs, both for classification and for regression. In this paper we analyze one popular approximation to GPs for regression: the reduced rank approximation. While generally GPs are equivalent to infinite linear models, we show that Reduced Rank Gaussian Processes (RRGPs) are equivalent to finite sparse linear models. We also introduce the concept of degenerate GPs and show that they correspond to inappropriate priors. We show how to modify the RRGP to prevent it from being degenerate at test time. Training RRGPs consists both in learning the covariance function hyperparameters and the support set. We propose a method for learning hyperparameters for a given support set. We also review the Sparse Greedy GP (SGGP) approximation (Smola and Bartlett, 2001), which is a way of learning the support set for given hyperparameters based on approximating the posterior. We propose an alternative method to the SGGP that has better generalization capabilities. Finally we make experiments to compare the different ways of training a RRGP. We provide some Matlab code for learning RRGPs.
    Original languageEnglish
    Title of host publicationSwitching and Learning in Feedback Systems
    Place of PublicationBerlin
    PublisherSpringer-verlag Berlin
    Publication date2005
    Publication statusPublished - 2005
    SeriesLecture Notes in Computer Science

    Fingerprint Dive into the research topics of 'Analysis of some methods for reduced rank Gaussian process regression'. Together they form a unique fingerprint.

    Cite this