
This Article  
 
Share  
Bibliographic References  
Add to:  
Digg Furl Spurl Blink Simpy Del.icio.us Y!MyWeb  
Search  
 
2009 WRI World Congress on Computer Science and Information Engineering
SieveDecrease Algorithms of Polynomial Neural Networks
Los Angeles, California USA
March 31April 02
ISBN: 9780769535074
ASCII Text  x  
Zou Ajin, Zhang Yunong, "SieveDecrease Algorithms of Polynomial Neural Networks," Computer Science and Information Engineering, World Congress on, vol. 5, pp. 564569, 2009 WRI World Congress on Computer Science and Information Engineering, 2009.  
BibTex  x  
@article{ 10.1109/CSIE.2009.128, author = {Zou Ajin and Zhang Yunong}, title = {SieveDecrease Algorithms of Polynomial Neural Networks}, journal ={Computer Science and Information Engineering, World Congress on}, volume = {5}, year = {2009}, isbn = {9780769535074}, pages = {564569}, doi = {http://doi.ieeecomputersociety.org/10.1109/CSIE.2009.128}, publisher = {IEEE Computer Society}, address = {Los Alamitos, CA, USA}, }  
RefWorks Procite/RefMan/Endnote  x  
TY  CONF JO  Computer Science and Information Engineering, World Congress on TI  SieveDecrease Algorithms of Polynomial Neural Networks SN  9780769535074 SP564 EP569 A1  Zou Ajin, A1  Zhang Yunong, PY  2009 KW  neural networks KW  polynomials KW  Sievedecrease KW  pseudoinverse VL  5 JA  Computer Science and Information Engineering, World Congress on ER   
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/CSIE.2009.128
To overcome the problem of determining the number of hiddenlayer neurons in feedforward neural networks, a polynomial feedforward neural network with a single hidden layer is presented based on the theory of polynomial approximation, where the polynomials are employed as the activation functions of hiddenlayer neurons, and the weights between input layer and hidden layer are set to be 1. We only need to adjust the weights between hidden layer and output layer. Then, using the least square method, we could deduce the formula of computing weights directly. Furthermore, the basic ideas of the sievedecrease algorithm of polynomial neural networks are described and discussed in details, together with several new concepts, such as weightsieve, sievepore diameter, sievedecrease rate,etc. Two illustrative computersimulations substantiate that the improved polynomial feedforward neural networks possess superior performance, and show that the number of hidden neurons decreases respectively by 98.19% and 80%, as compared to primal neural networks.
Index Terms:
neural networks, polynomials, Sievedecrease, pseudoinverse
Citation:
Zou Ajin, Zhang Yunong, "SieveDecrease Algorithms of Polynomial Neural Networks," csie, vol. 5, pp.564569, 2009 WRI World Congress on Computer Science and Information Engineering, 2009
Usage of this product signifies your acceptance of the Terms of Use.