Genetic Algorithm for Neurocomputer Image Recognition
Genetic algorithms for optimization of feature set and internal structure of neural networks are considered. Results of experimental investigation of genetic algorithms are given. Experiments show that performance of neural networks after such optimization substantially increases.
KeywordsGenetic Algorithm Cost Function Input Structure Error Number Handwritten Word
Unable to display preview. Download preview PDF.
- 2.Hecht-Nielsen, R. Neurocomputing. Addison-Wesley, Reading, MA, 1990.Google Scholar
- 3.Kussul, E.M., Luk, A.N. Soviet science review. Scientific developments in the USSR. 3 (3), 168 (1972).Google Scholar
- 6.Kussul, E.M., Baidyk, T.N., Lukovitch, V.V., Rachkovskij, D.A. Adaptive high performance classifier based on random threshold neurons. Proc. of Twelfth European Meeting on Cybernetics and Systems Research (EMCSR-94), Austria, Vienna, April 5–8, 1994 in R. Trappl (ed.): Cybernetics and Systems’94, World Scientific Publishing Co.Pte.Ltd, Singapore. - P. 1687–1695.Google Scholar
- 7.Kussul, E.M., Baidyk, T.N. Neural Random Threshold Classifier in OCR Application. Proc. of the Second All-Ukrainian Intern. Conf.“UkrO- BRAZ’94”, Kyjiv, Ukraine, December 20–24, 1994. - P. 154–157.Google Scholar
- 8.Kussul, E.M., Rachkovskij, D.A., Baidyk, T.N. Associative-projective neural networks: architecture, implementation, applications. Proc. of Fourth Intern. Conf. “Neural Networks & their Applications”, Nimes, France, Nov.4–8 1991. - P. 463–476.Google Scholar