On the generalization problem

DOI10.1023/A:1012693710097MaRDI QIDQ5956967zbMATH OpenOpenAlexFDO

Authors L. C. W. Dixon

Publication date 26 February 2002

Published in Journal of Optimization Theory and Applications (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1023/a:1012693710097

zbMATH Keywords

constrained nonlinear optimizationnetworks

Mathematics Subject Classification ID

Approximation by other special function classes (41A30)

Review text

The generalization problem consists of finding the type of networks of which the output \(y\) will approximate the output \(s(x)\) of a given system for any value of the system input \(x\). Three cases are considered: (a) \((0,1)\) output; (b) accurate real output and (c) noisy real output. It is shown that it is necessary to have prior knowledge about the output of the system, to solve the problem accurately(?). In the case (c), the least squares approach works badly, and a regularization method is proposed to circumvent this defect. The cases (a) and (b) lead to constrained nonlinear optimization problems.

Recommendations

• scientific article; zbMATH DE number 6196
• scientific article; zbMATH DE number 1843046
• scientific article; zbMATH DE number 1322247
• scientific article; zbMATH DE number 4211541
• Neural Network Learning as an Inverse Problem

Cites work

• Approximation by superpositions of a sigmoidal function
• Geometric approach to Fletcher's ideal penalty function
• scientific article; zbMATH DE number 3551792 (Why is no real title available?)
• Multilayer feedforward networks are universal approximators

Cited in

(13)

• Improved generalization via tolerant training
• Generality's price: Inescapable deficiencies in machine-learned programs
• Investigation of generalization ability of a trained stochastic kinetic model of neuron
• On the Problem in Model Selection of Neural Network Regression in Overrealizable Scenario
• scientific article; zbMATH DE number 3843167 (Why is no real title available?)
• Generality in artificial intelligence
• scientific article; zbMATH DE number 96338 (Why is no real title available?)
• scientific article; zbMATH DE number 1159251 (Why is no real title available?)
• scientific article; zbMATH DE number 1961159 (Why is no real title available?)
• scientific article; zbMATH DE number 6196 (Why is no real title available?)
• LEARNING DESIGN KNOWLEDGE AS GENERALIZATION
• Generalization improvement of wavelet neural networks using regularization techniques
• Neural Network Learning as an Inverse Problem

This page was built for publication: On the generalization problem

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5956967)