Learning Curves for Gaussian Processes
Abstract
I consider the problem of calculating learning curves (i.e., average generalization performance) of Gaussian processes used for regres(cid:173) sion. A simple expression for the generalization error in terms of the eigenvalue decomposition of the covariance function is derived, and used as the starting point for several approximation schemes. I identify where these become exact, and compare with existing bounds on learning curves; the new approximations, which can be used for any input space dimension, generally get substantially closer to the truth.
Cite
Text
Sollich. "Learning Curves for Gaussian Processes." Neural Information Processing Systems, 1998.Markdown
[Sollich. "Learning Curves for Gaussian Processes." Neural Information Processing Systems, 1998.](https://mlanthology.org/neurips/1998/sollich1998neurips-learning/)BibTeX
@inproceedings{sollich1998neurips-learning,
title = {{Learning Curves for Gaussian Processes}},
author = {Sollich, Peter},
booktitle = {Neural Information Processing Systems},
year = {1998},
pages = {344-350},
url = {https://mlanthology.org/neurips/1998/sollich1998neurips-learning/}
}