The improved localized generalization error model and its applications to feature selection for RBFNN
- Resource Type
- Conference
- Authors
- Cui, Yan-Jun; Li, Jie; Ma, Yan-Dong
- Source
- 2010 International Conference on Machine Learning and Cybernetics Machine Learning and Cybernetics (ICMLC), 2010 International Conference on. 3:1515-1518 Jul, 2010
- Subject
- Computing and Processing
Communication, Networking and Broadcast Technologies
Training
Computational modeling
Accuracy
Machine learning
Glass
Cybernetics
Iris
Localization Generalization Error
l-norm
Radial Basis Function Neural Networks (RNFNN)
Feature Selection
- Language
- ISSN
- 2160-133X
2160-1348
In pattern classification problems, the generalization error caused more and more attentions because of its importance for classifier's training. Wing W.Y. NG [1] et al. proposed localized generalization error model compared to global generalization error model. The idea is perfect, but the derivation of the error model and stochastic sensitivity measure has some flaws. In this paper, we propose an improved localized generalization error model in order to avoid these flaws of the model proposed by Wing.