Neuron-adaptive higher order neural network group models

Author

Zhang, Ming ; Xu, Shuxiang ; Lu, Bo

Author_Institution

Dept. of Comput. & Inf. Syst., Western Sydney Univ., Campbelltown, NSW, Australia

Volume

1

fYear

1999

fDate

6/21/1905 12:00:00 AM

Firstpage

333

Abstract

Neuron-adaptive higher order neural network group (NAHONG) models with a neuron-adaptive activation function are studied in this paper. A learning algorithm is derived to adjust free parameters in the neuron-adaptive activation function as well as connection weights. Our experiments had confirmed that NAHONG model holds superior approximation capabilities to traditional neuron-fixed neural network group model and normal neuron-adaptive neural network group model. Our experiments also reveal that NAHONG is especially exceptional in simulating complex piecewise functions

Keywords

adaptive systems; learning (artificial intelligence); modelling; neural nets; transfer functions; NAHONG models; approximation capabilities; complex piecewise function simulation; connection weights; free parameter adjustment; neuron-adaptive activation function; neuron-adaptive high-order neural network group models; Adaptive systems; Computational modeling; Computer networks; Feedforward neural networks; Information systems; Multi-layer neural network; Neural networks; Neurons; Pattern recognition; Roentgenium;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks, 1999. IJCNN '99. International Joint Conference on

ISSN

1098-7576

Print_ISBN

0-7803-5529-6

Type

conf

DOI

10.1109/IJCNN.1999.831513

Filename

831513