Using the Equivalent Kernel to Understand Gaussian Process Regression

@inproceedings{Sollich2004UsingTE,
  title={Using the Equivalent Kernel to Understand Gaussian Process Regression},
  author={Peter Sollich and Christopher K. I. Williams},
  booktitle={NIPS},
  year={2004}
}
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. Consider the supervised regression problem for a dataset D with entries (xi, yi) for i = 1… CONTINUE READING

Citations

Publications citing this paper.
SHOWING 1-10 OF 28 CITATIONS, ESTIMATED 23% COVERAGE

124 Citations

01020'07'10'13'16'19
Citations per Year
Semantic Scholar estimates that this publication has 124 citations based on the available data.

See our FAQ for additional information.

References

Publications referenced by this paper.
SHOWING 1-10 OF 12 REFERENCES

Annals of Statistics

  • B. W. Silverman
  • 12:898–916
  • 1984
Highly Influential
6 Excerpts

In NIPS 14

  • P. Sollich
  • pages 519–526
  • 2002
1 Excerpt

Neural Computation

  • P. Sollich, A. Halees
  • 14:1393–1428
  • 2002

In NIPS 11

  • M. Opper, F. Vivarelli
  • pages 302–308
  • 1999
2 Excerpts

In M

  • C.K.I. Williams
  • I. Jordan, editor, Learning in Graphical Models…
  • 1998

Numerical Algorithms

  • C. Thomas-Agnan
  • 13:21–32
  • 1996
1 Excerpt

Neural Computation

  • F. Girosi, M. Jones, T. Poggio
  • 7(2):219–269
  • 1995
3 Excerpts

Probability

  • A. Papoulis
  • Random Variables, and Stochastic Processes…
  • 1991

Similar Papers

Loading similar papers…