Using the equivalent kernel to understand Gaussian process regression

Peter Sollich; Christopher K.I. Williams

Using the equivalent kernel to understand Gaussian process regression

Peter Sollich, Christopher K.I. Williams

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

The equivalent kernel [1] is a way of understanding how Gaussian process regression works for large sample sizes based on a continuum limit. In this paper we show (1) how to approximate the equivalent kernel of the widely-used squared exponential (or Gaussian) kernel and related kernels, and (2) how analysis using the equivalent kernel helps to understand the learning curves for Gaussian processes.

Original language	English
Title of host publication	Advances in Neural Information Processing Systems 17 (NIPS 2004)
Pages	1313-1320
Number of pages	8
Publication status	Published - 2004

Access to Document

2676-using-the-equivalent-kernel-to-understand-gaussian-process-regressionFinal published version, 299 KB

http://papers.nips.cc/book/advances-in-neural-information-processing-systems-17-2004

Cite this

@inproceedings{daf826e567214281980061bffd41f827,

title = "Using the equivalent kernel to understand Gaussian process regression",

abstract = "The equivalent kernel [1] is a way of understanding how Gaussian process regression works for large sample sizes based on a continuum limit. In this paper we show (1) how to approximate the equivalent kernel of the widely-used squared exponential (or Gaussian) kernel and related kernels, and (2) how analysis using the equivalent kernel helps to understand the learning curves for Gaussian processes.",

author = "Peter Sollich and Williams, {Christopher K.I.}",

year = "2004",

language = "English",

pages = "1313--1320",

booktitle = "Advances in Neural Information Processing Systems 17 (NIPS 2004)",

}

TY - GEN

T1 - Using the equivalent kernel to understand Gaussian process regression

AU - Sollich, Peter

AU - Williams, Christopher K.I.

PY - 2004

Y1 - 2004

N2 - The equivalent kernel [1] is a way of understanding how Gaussian process regression works for large sample sizes based on a continuum limit. In this paper we show (1) how to approximate the equivalent kernel of the widely-used squared exponential (or Gaussian) kernel and related kernels, and (2) how analysis using the equivalent kernel helps to understand the learning curves for Gaussian processes.

AB - The equivalent kernel [1] is a way of understanding how Gaussian process regression works for large sample sizes based on a continuum limit. In this paper we show (1) how to approximate the equivalent kernel of the widely-used squared exponential (or Gaussian) kernel and related kernels, and (2) how analysis using the equivalent kernel helps to understand the learning curves for Gaussian processes.

M3 - Conference contribution

SP - 1313

EP - 1320

BT - Advances in Neural Information Processing Systems 17 (NIPS 2004)

ER -