Title :
The importance of maintaining behavioural link between parents and offspring
Author_Institution :
Comput. Intelligence Group, Australian Defence Force Acad., Canberra, ACT, Australia
Abstract :
In a study of evolutionary artificial neural networks (Yao and Liu, 1996), it has been argued that a partial training process after an architectural mutation plays an important role in maintaining the behavioural link between parents and their offspring and thus is beneficial to the simulated evolution. This paper investigates the issue further through a number of experiments. The experimental results show that a closer behavioural link between parents and their offspring due to the partial training process does lead to better performance, i.e., evolved ANNs generalise better. The results also illustrate that given a fixed amount of time there is an optimal balance of time between evolution and training (learning)
Keywords :
generalisation (artificial intelligence); genetic algorithms; genetics; learning (artificial intelligence); neural nets; architectural mutation; behavioural link; evolutionary artificial neural networks; experiments; generalisation; learning; offspring; parents; partial training process; performance; simulated evolution; time; Artificial neural networks; Australia; Computational intelligence; Computer science; Educational institutions; Genetic mutations; Genetic programming; Springs;
Conference_Titel :
Evolutionary Computation, 1997., IEEE International Conference on
Conference_Location :
Indianapolis, IN
Print_ISBN :
0-7803-3949-5
DOI :
10.1109/ICEC.1997.592388