Whitley, D., & Hanson, T. (1989). Towards the genetic synthesis of neural networks.  Home/Search Document Not in Database Summary Related Articles Check |
This paper is cited in the following contexts:
The Design and Evolution of Modular Neural Network Architectures - Happel, Murre (1994) (33 citations) (Correct)
....In general, strings are randomly selected, with high fitness strings having a higher chance of being selected. For example, the probability of a string to be selected can be taken proportional to its absolute fitness value (e.g. Goldberg, 1989) Another possibility is rank based selection (Whitley, 1989), in which the selection probability is defined as a linear function of the rank of a string in the population according to its fitness. With all simulations in this study, the static population model (Whitley, 1989) will be used for selection and replacement of solutions. Apart from rank based ....
....fitness value (e.g. Goldberg, 1989) Another possibility is rank based selection (Whitley, 1989) in which the selection probability is defined as a linear function of the rank of a string in the population according to its fitness. With all simulations in this study, the static population model (Whitley, 1989) will be used for selection and replacement of solutions. Apart from rank based selection this model uses one at a time selection and replacement. A new solution is inserted at the appropriate position (rank) Happel and Murre The design and evolution of modular neural networks 19 and pushes ....
[Article contains additional citation context not shown here]
Whitley, D., & Hanson, T. (1989). Towards the genetic synthesis of neural networks.
Designing Modular Artificial Neural Networks - Boers, Kuiper, Happel.. (1993) (18 citations) (Correct)
....crossover, inversion and mutation. From generation to generation the mean fitness of the population should increase. Genetic algorithms have proved to provide a powerful method for solving multiple constraint problems. The specific genetic algorithm used is derived from the GENITOR algorithm [26, 27] that uses a static population model. In this model, that uses rank based selection, each new member of the population is inserted in the population according to its fitness, removing the population s worst member and keeping the population sorted according to fitness. 3.1 Coding the production ....
D. Whitley and T. Hanson; `Towards the genetic synthesis of neural networks'. In: Proceedings of the 3rd International Conference on Genetic Algorithms and their applications (ICGA), 391--396, J.D. Schaffer (Ed.), Morgan Kaufmann, San Mateo CA, 1989.
Biological Metaphors and the Design of Modular Artificial.. - Boers, Kuiper (1992) (46 citations) (Correct)
....algorithm like backpropagation. The genes of the algorithm have a one to one correspondence to the weights of the network. A slight variation of this method is to use the genetic algorithm to find a set of reasonably good weights, leaving the fine tuning to a learning algorithm (see for example [WHIT89B] and [GARI90] Use the genetic algorithm to find the structure of a network. With this method the genetic algorithm tries to find the optimal structure of a network, instead of the weights of a given structure. The genes of the genetic algorithm now contain a coding for the topology of the ....
D. Whitley and T. Hanson; `Towards the genetic synthesis of neural networks'. In: Proceedings of the 3rd International Conference on Genetic Algorithms and their applications (ICGA), 391-396, J.D. Schaffer (Ed.), Morgan Kaufmann, San Mateo CA, 1989.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC