DocumentCode
2837612
Title
An ANN model of optimizing activation functions based on constructive algorithm and GP
Author
Sheng, Zhang ; Xiuyu, Shang ; Wei, Wang
Author_Institution
Coll. of Math., Phys. & Inf. Eng., Zhejiang Normal Univ., Jinhua, China
Volume
1
fYear
2010
fDate
22-24 Oct. 2010
Abstract
The importance of the activation functions in ANN is emphasized. A new ANN modeling method is proposed based on constructive algorithm and GP. This method can be used to realize the automatic optimization of the ANN´s net structure and the activation functions. As a result, the ANN´s constructure and generalization capability is greatly improved, it´s characteristic is better than the M-P feed forword neural network. This improvement is verified experimentally.
Keywords
genetic algorithms; learning (artificial intelligence); neural nets; ANN activation functions; ANN modeling method; artificial neural nets; constructive algorithm; genetic programming; Artificial neural networks; Bayesian methods; Neurons; Training; ANN; Activation Functions; Constructive Algorithm; GP;
fLanguage
English
Publisher
ieee
Conference_Titel
Computer Application and System Modeling (ICCASM), 2010 International Conference on
Conference_Location
Taiyuan
Print_ISBN
978-1-4244-7235-2
Electronic_ISBN
978-1-4244-7237-6
Type
conf
DOI
10.1109/ICCASM.2010.5620620
Filename
5620620
Link To Document