Title :
Beyond weights adaptation: a new neuron model with trainable activation function and its supervised learning
Author :
Wu, Youshou ; Zhao, Mingsheng ; Ding, Xiaoqing
Author_Institution :
Dept. of Electron. Eng., Tsinghua Univ., Beijing, China
Abstract :
This paper proposes a new kind of neuron model, which has trainable activation function (TAF) in addition to only trainable weights in the conventional M-P model. The final neuron activation function can be derived by training a primitive neuron activation function. BP like learning algorithm has been presented for MFNN constructed by neurons of TAF model. Two simulation examples are given to show the network capacity and performance advantages of the new MFNN in comparison with that of conventional sigmoid MFNN
Keywords :
backpropagation; feedforward neural nets; multilayer perceptrons; BP-like learning algorithm; M-P model; MFNN; TAF; backpropagation; multilayer feedforward neural net; neuron model; primitive neuron activation function training; supervised learning; trainable activation function; weight adaptation; Artificial neural networks; Biological neural networks; Extraterrestrial measurements; Image processing; Nervous system; Neural networks; Neurons; Performance analysis; Research and development; Supervised learning;
Conference_Titel :
Neural Networks,1997., International Conference on
Conference_Location :
Houston, TX
Print_ISBN :
0-7803-4122-8
DOI :
10.1109/ICNN.1997.616194
