| N. Dodd. Optimization of neural network structures using genetic techniques. Internal Report RIPREP/1000/63/89, Royal Signals and Radar Establishment, Malvern, UK, 1989. |
....is invariably going to be one, two or (at most three) However, given that we have (arbitrarily) set the number of hidden layers it might be worthwhile to use GAs to determine the number of nodes in each layer. There have been several papers using GAs to design the overall topology of a network [Dodd 1989], Miller 1989] Whitley 1989] Whitley 1990] Whitley 1991] Dodd 1991] and these have yeilded interesting results on the optimisation of network design. They have not, however, led to qualitatively new kinds of connectionist processes. Moreover, functionally equivalent networks with ....
....to neural networks. Of particular interest are the representations selected and the extent to which the crossover operator designed for the representation is liable to preserve structurally coadapted building blocks. In some cases the authors commented on the problem of permutational redundancy. Dodd [Dodd 1989] applied a GA to optimise a structured network for a pattern recognition problem classifying dolphin sounds. The network was called a spread network and consisted of a two dimensional grid of input nodes, an input retina , and a number of hidden nodes connected to rectangular patches on the ....
N. Dodd. Optimization of neural network structures using genetic techniques. Internal Report RIPREP/1000/63/89, Royal Signals and Radar Establishment, Malvern, UK, 1989.
No context found.
N. Dodd, "Optimization of neural-network structure using genetic techniques," in Proc. Conf. Applications of Artificial Intelligence in Engineering VI, G. Rzevski and R. A. Adey, Eds. London, U.K.: Elsevier, 1991, pp. 939--944.
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