Title :
Parametrical synthesis of neural network models based on the evolutionary optimization
Author :
Oleynik, Alexey ; Subbotin, Sergey
Author_Institution :
Dept. of Software, Zaporozhye Nat. Tech. Univ., Zaporozhye, Ukraine
Abstract :
The problem of parametrical synthesis of neural network models is considered. The evolutionary method with using the aprioristic information for neural network training is developed. Experiments on neural models synthesis for medical diagnostics are lead.
Keywords :
evolutionary computation; neural nets; neurophysiology; evolutionary optimization; medical diagnostics; neural network models; neural network training; parametrical synthesis; Biological cells; Genetic mutations; Gradient methods; Management training; Network synthesis; Neural networks; Neurons; Optimization methods; Stochastic processes; Transfer functions; Evolutionary optimization; Model; Neural Network;
Conference_Titel :
CAD Systems in Microelectronics, 2009. CADSM 2009. 10th International Conference - The Experience of Designing and Application of
Conference_Location :
Lviv-Polyana
Print_ISBN :
978-966-2191-05-9
