Upper and Lower Bounds on the Learning Curve for Gaussian Processes

Authors:
Christopher K. I. Williams;Francesco Vivarelli
Affiliations:
Division of Informatics, The University of Edinburgh, 5 Forrest Hill, Edinburgh EH1 2QL, Scotland, UK. ckiw@dai.ed.ac.uk;Neural Computing Research Group, Department of Computer Science and Applied Mathematics, Aston University, Birmingham B4 7ET, UK. vivarelf@aston.ac.uk
Venue:
Machine Learning
Year:
2000

Citing 9
Cited 4

A theory of the learnable

Communications of the ACM
Stochastic linear learning: exact test and training error averages

Neural Networks
The nature of statistical learning theory

The nature of statistical learning theory
Almost optimal differentiation using noisy data

Journal of Approximation Theory
An upper bound on the Bayesian error bars for generalized linear regression

MANNA '95 Proceedings of the first international conference on Mathematics of neural networks : models, algorithms and applications: models, algorithms and applications
General bounds on Bayes errors for regression with Gaussian processes

Proceedings of the 1998 conference on Advances in neural information processing systems II
Learning curves for Gaussian processes

Proceedings of the 1998 conference on Advances in neural information processing systems II
Bayesian Learning for Neural Networks

Bayesian Learning for Neural Networks
Evaluation of gaussian processes and other methods for non-linear regression

Evaluation of gaussian processes and other methods for non-linear regression

Learning curves for Gaussian process regression: approximations and bounds

Neural Computation
Learning curves for Gaussian processes via numerical cubature integration

ICANN'11 Proceedings of the 21th international conference on Artificial neural networks - Volume Part I
Can gaussian process regression be made robust against model mismatch?

Proceedings of the First international conference on Deterministic and Statistical Methods in Machine Learning
Random walk kernels and learning curves for Gaussian process regression on random graphs

The Journal of Machine Learning Research

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper we introduce and illustrate non-trivial upper and lower bounds on the learning curves for one-dimensional Guassian Processes. The analysis is carried out emphasising the effects induced on the bounds by the smoothness of the random process described by the Modified Bessel and the Squared Exponential covariance functions. We present an explanation of the early, linearly-decreasing behavior of the learning curves and the bounds as well as a study of the asymptotic behavior of the curves. The effects of the noise level and the lengthscale on the tightness of the bounds are also discussed.