Using One-Dimensional Linear Interpolation Method to Check Over-Fitting in Neural Network with Multi-Dimensional Inputs

Bao Lien Hung; Hsin An Hung

doi:10.4028/www.scientific.net/AMM.496-500.2228

Paper Titles

Current Status and Development Trend of Research of Computer Forensic
p.2208

Infinitely many Solutions of Semi-Linear Elliptic Equation with a Logarithmic Nonlinear Term
p.2216

Operating Mechanism of Teaching Quality Information System
p.2220

Design of Polytechnic College Practice Training Base Management System Based on Internet of Things
p.2224

Using One-Dimensional Linear Interpolation Method to Check Over-Fitting in Neural Network with Multi-Dimensional Inputs
p.2228

Schur Factorization for Unitary Extended Matrix
p.2233

Image Tamper Blind Detection Based on CFA
p.2237

A New Automatic Seamless Image Stitching Algorithm Based on the Gray Value of Edges
p.2241

A Simplified Integration for Heterogeneous Database Application System
p.2246

HomeApplied Mechanics and MaterialsApplied Mechanics and Materials Vols. 496-500Using One-Dimensional Linear Interpolation Method...

Using One-Dimensional Linear Interpolation Method to Check Over-Fitting in Neural Network with Multi-Dimensional Inputs

Abstract:

Neural networks are widely used to learn and predict the correlation between input and output. However, in the process of learning, the excessive reduction of errors may influence the validity of prediction, this phenomenon is called over-fitting. In order to address this problem, this study sequenced the input data into one-dimensional data for the neural network structure of multi-dimensional inputs, and used visual graphics to observe whether there is over-fitting. This method is called one-dimensional linear interpolation method. The result of example validation proved that the proposed method can provide specific graphical information effectively, and determine whether there is over-fitting.

You might also be interested in these eBooks

Frontiers of Manufacturing and Design Science IV

View Preview

Info:

Periodical:

Applied Mechanics and Materials (Volumes 496-500)

Pages:

2228-2232

DOI:

https://doi.org/10.4028/www.scientific.net/AMM.496-500.2228

Citation:

Cite this paper

Online since:

January 2014

Authors:

Bao Lien Hung, Hsin An Hung*

Keywords:

Multi-Dimensional Input, Neural Network (NN), One-Dimensional Linear Interpolation, Over-Fitting, Visualization

Export:

RIS, BibTeX

Price:

Permissions CCC:

Request Permissions

Permissions PLS:

Request Permissions

Сopyright:

Citation:

* - Corresponding Author

References

[1] A. Krogh and J.A. Hertz, in: In Advances in Neural Information Processing Systems 4, edited by J.E. Moody, S.J. Hanson and R.P. Lippmann, p.950, Morgan Kaumann Publishers (1992).

Google Scholar

[2] L. Prechelt: Neural Networks, Vol. 11, Issue 04 (1998), p.761.

Google Scholar

[3] Z.L. Wang, H.R. Li and S.S. Gu: Journal of Northeastern University (Natural Science), Vol. 22, No. 4 (2001), p.260.

Google Scholar

[4] J.C. Li, G.J. Qin, X.S. Wen and N.Q. Hu: Journal of Vibration, Measurtment & Diagnosis, Vol. 22, No. 4 (2002), p.358.

Google Scholar

[5] G. Panchal, A. Ganatra, P. Shah, and D. Panchal: Int. J. Soft Computing, Vol. 2, No. 2 (2011), p.40.

Google Scholar

[6] H.A. Hung and H.T. Hsiao: Journal of Chienkuo Technology University, Vol. 20 (2001), p.243.

Google Scholar

[7] H. A. Hung and C. C. Hsieh: The 29th National Conference of CSME (2012).

Google Scholar