Confidence intervals for the risks of regression models

Authors:
Imhoi Koo;Rhee Man Kil
Affiliations:
Division of Applied Mathematics, Korea Advanced Institute of Science and Technology, Daejeon, Korea;Division of Applied Mathematics, Korea Advanced Institute of Science and Technology, Daejeon, Korea
Venue:
ICONIP'06 Proceedings of the 13 international conference on Neural Information Processing - Volume Part I
Year:
2006

Citing 4
Cited 1

Layered neural networks with Gaussian hidden units as universal approximations

Neural Computation
A Gaussian potential function network with hierarchically self-organizing learning

Neural Networks
Polynomial bounds for VC dimension of sigmoidal neural networks

STOC '95 Proceedings of the twenty-seventh annual ACM symposium on Theory of computing
Fast learning in networks of locally-tuned processing units

Neural Computation

Parameterized cross-validation for nonlinear regression models

Neurocomputing

Quantified Score

Hi-index	0.00

Visualization

Abstract

The empirical risks of regression models are not accurate since they are evaluated from the finite number of samples. In this context, we investigate the confidence intervals for the risks of regression models, that is, the intervals between the expected and empirical risks. The suggested method of estimating confidence intervals can provide a tool for predicting the performance of regression models.