Title :
The analysis of convergence of hybrid algorithm based on Neural Network and Genetic Algorithm
Author :
PAN, Mei-Qin ; HE, Guo-Ping
Author_Institution :
Coll. of Inf. Sci. & Eng., Shan Dong Univ. of Sci. & Technol., Qingdao
Abstract :
This paper analyzes the advantages and disadvantages of GA and BP algorithms, and presents the iteration of hybrid algorithm based on both algorithms. The hybrid algorithm incorporates the stronger global search of GA into the stronger local search of BP, and can search out the global optimum faster than each algorithm. At last, the hybrid algorithm is proved converge to the global optimum with the probability of 1
Keywords :
convergence of numerical methods; genetic algorithms; neural nets; backpropagation algorithms; convergence analysis; genetic algorithm; hybrid algorithm; neural network; Algorithm design and analysis; Biological neural networks; Convergence; Educational institutions; Genetic algorithms; Genetic mutations; Iterative algorithms; Neural networks; Paper technology; Signal processing algorithms;
Conference_Titel :
Neural Networks and Brain, 2005. ICNN&B '05. International Conference on
Conference_Location :
Beijing
Print_ISBN :
0-7803-9422-4
DOI :
10.1109/ICNNB.2005.1614604
