A neuron model with trainable activation function (TAF) and its MFNN supervised learning

作     者：吴佑寿 赵明生 

作者机构：1. Department of Electronic Engineering Tsinghua University 100084 Beijing China 

出 版 物：《Science in China(Series F)》 (中国科学（F辑英文版）)

年 卷 期：2001年第44卷第5期

页      面：366-375页

学科分类：12[管理学] 1201[管理学-管理科学与工程(可授管理学、工学学位)] 081104[工学-模式识别与智能系统] 08[工学] 0835[工学-软件工程] 0811[工学-控制科学与工程] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：This work was supported by the National Natural Science Foundation of China (Grant Nos. 69831030 and 630003014) 

主　　题：neuron model neural network TAF neuron model learning algorithm. 

摘      要：This paper addresses 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 from a primitive neuron activation function by training. The BP like learning al-gorithm has been presented for MFNN constructed by neurons of TAP model. Several simulation ex-amples are given to show the network capacity and performance advantages of the new MFNN in com-parison with that of conventional sigmoid MFNN.