Generalization Error Estimation for Non-linear Learning Methods

Masashi SUGIYAMA  

IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences   Vol.E90-A   No.7   pp.1496-1499
Publication Date: 2007/07/01
Online ISSN: 1745-1337
DOI: 10.1093/ietfec/e90-a.7.1496
Print ISSN: 0916-8508
Type of Manuscript: LETTER
Category: Neural Networks and Bioengineering
supervised learning,  generalization capability,  model selection,  subspace information criterion,  non-linear learning,  

Full Text: PDF>>
Buy this Article

Estimating the generalization error is one of the key ingredients of supervised learning since a good generalization error estimator can be used for model selection. An unbiased generalization error estimator called the subspace information criterion (SIC) is shown to be useful for model selection, but its range of application is limited to linear learning methods. In this paper, we extend SIC to be applicable to non-linear learning.