Title :
Optimizing Weight and Threshold of BP Neural Network Using SFLA: Applications to Nonlinear Function Fitting
Author :
Hongwei Ye ; Linfang Yang ; Xiaozhang Liu
Author_Institution :
Sch. of Electron. & Inf. Eng., Heyuan Polytech., Heyuan, China
Abstract :
The shuffled frog-leaping algorithm (SFLA) is presented along with a pseudocode and flow chart to facilitate its implementation. And then we use SFLA to optimize the weight and the threshold value of BP network. Based on the experiments, we show that SFLA performs better than Genetic Algorithms (GAs) in the optimization of BP network´s weight and threshold value, which are used in the nonlinear function fitting.
Keywords :
backpropagation; mathematics computing; neural nets; nonlinear functions; optimisation; BP neural network optimizing weight; BP neural network threshold; SFLA; flow chart; nonlinear function fitting; pseudocode; shuffled frog-leaping algorithm; Algorithm design and analysis; Fitting; Neural networks; Prediction algorithms; Sociology; Statistics; Training; BP network; SFLA; nonlinear function fitting;
Conference_Titel :
Emerging Intelligent Data and Web Technologies (EIDWT), 2013 Fourth International Conference on
Conference_Location :
Xi´an
Print_ISBN :
978-1-4799-2140-9
DOI :
10.1109/EIDWT.2013.41
