J4 ›› 2013, Vol. 35 ›› Issue (5): 142-148.
• 论文 • Previous Articles Next Articles
ZHANG Yunong,LUO Feiheng,CHEN Jinhao,LI Weibing
Received:
Revised:
Online:
Published:
Abstract:
Based on the function approximation theory and the Weierstrass approximation theorem, a novel 3input neural net activated by a group of products of Bernoulli polynomials (i.e., 3input Bernoulli polynomial neural net, 3IBPNN) was constructed in this paper. Furthermore, on the basis of the weightsdirectdetermination (WDD) method and the relationship between the number of hiddenlayer neurons and the approximation error of the neural net, three different weightsandstructuredetermination (WASD) algorithms were built up for the constructed 3IBPNN. Numerical experiment results further prove that all of the 3IBPNNs determined respectively by the three proposed algorithms perform excellently in training, testing and prediction.
Key words: Bernoulli polynomial neural net;weightsdirectdetermination method;weightsandstructuredetermination algorithm;algorithm;numerical experiment
ZHANG Yunong,LUO Feiheng,CHEN Jinhao,LI Weibing. Weights and structure determination of 3-input Bernoulli polynomial neural net[J]. J4, 2013, 35(5): 142-148.
0 / / Recommend
Add to citation manager EndNote|Ris|BibTeX
URL: http://joces.nudt.edu.cn/EN/
http://joces.nudt.edu.cn/EN/Y2013/V35/I5/142