Title :
A mixed genetic approach to the optimization of neural controllers
Author :
Heistermann, Jochen
Author_Institution :
Siemens AG, Munich, Germany
Abstract :
The author discusses some of the capabilities of genetic algorithms (GAs). GAs are compared with other standard optimization methods like gradient descent or simulated annealing (SA). It is shown that SA is just a special case of GA. The role of a population in the optimization process is demonstrated by an example. GA was applied as a learning algorithm to neural networks.<>
Keywords :
genetic algorithms; learning (artificial intelligence); neural nets; simulated annealing; genetic algorithms; gradient descent; learning algorithm; neural networks; optimization methods; simulated annealing; Artificial neural networks; Biological neural networks; Genetic algorithms; Humans; Learning; Neural networks; Optimization methods; Pattern recognition; Research and development; Simulated annealing;
Conference_Titel :
CompEuro '92 . 'Computer Systems and Software Engineering',Proceedings.
Conference_Location :
The Hague, Netherlands
Print_ISBN :
0-8186-2760-3
DOI :
10.1109/CMPEUR.1992.218440
