DocumentCode :
1991240
Title :
Using genetic algorithm with adaptive mutation mechanism for neural networks design and training
Author :
Tsoy, Y.R. ; Spitsyn, V.G.
Author_Institution :
Dept. of Comput. Eng., Tomsk Polytech. Univ., Russia
fYear :
2005
fDate :
26 June-2 July 2005
Firstpage :
709
Lastpage :
714
Abstract :
In this paper a developed evolutionary algorithm (NEvA) for simultaneous connections and weights of neural network training is described. A distinctive feature of the algorithm is flexible and effective evolutionary search and a balanced resulting neural network structure due to adaptive mutation operator. In NEvA neural network structure changes, caused by mutation operator, as well as mutation rate are defined independently for each individual. Two different problems are chosen to test the algorithm. The first one is a simple 2-bit parity problem, well known as XOR problem, and the second is a neurocontrol problem of 1 and 2 poles balancing. A comparison of obtained results with results of other algorithms is presented.
Keywords :
adaptive systems; genetic algorithms; neural nets; neurocontrollers; poles and zeros; search problems; NEvA neural network structure; XOR problem; adaptive mutation operator; evolutionary algorithm; evolutionary search; genetic algorithm; neural network design; neural network training; neurocontrol problem; neuroevolutionay algorithm; parity problem; pole balancing; Algorithm design and analysis; Artificial neural networks; Computer networks; Encoding; Evolutionary computation; Genetic algorithms; Genetic mutations; Neural networks; Neurons; Organisms;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Science and Technology, 2005. KORUS 2005. Proceedings. The 9th Russian-Korean International Symposium on
Print_ISBN :
0-7803-8943-3
Type :
conf
DOI :
10.1109/KORUS.2005.1507882
Filename :
1507882
Link To Document :
بازگشت