Title :
BG: a hybrid algorithm for MLP
Author :
Lu, Chun ; Shi, Bingxue ; Chen, Lu
Author_Institution :
Inst. of Microelectron., Tsinghua Univ., Beijing, China
Abstract :
The proposed hybrid BG learning method for a Multilayer Feedforward Perceptron (MLP) blends the merits of both the Back-Propagation algorithm (BP) and the Genetic Algorithm (GA). BG has superior performance over both BP and GA because it learns from their strong points to offset their weakness. Results of the experiments including the XOR problem and the sin(x) function approximation have proved the good performance of the BG algorithm
Keywords :
feedforward neural nets; function approximation; learning (artificial intelligence); multilayer perceptrons; MLP; XOR problem; hybrid BG learning method; multilayer feedforward perceptron; sin(x) function approximation; Convergence; Genetic mutations; Hardware; Neural networks;
Conference_Titel :
Circuits and Systems, 2000. IEEE APCCAS 2000. The 2000 IEEE Asia-Pacific Conference on
Conference_Location :
Tianjin
Print_ISBN :
0-7803-6253-5
DOI :
10.1109/APCCAS.2000.913509
