JF Latreille <s21371 AT rmc PISTE ca> writes:
> If someone studied the application of GA on neuro-networks, I would like
> to get more information on this.
One resource:
http://www.utu.fi/~magi/opinnot/gradu/
It contains also some references to other studies on evolutionary ANNs.
Title and abstract:
Marko Grönroos. 1998. Evolutionary Design of Neural Networks.
Master's thesis. Computer Science, Department of Mathematical
Sciences, University of Turku, Finland, 1998.
Abstract
This thesis deals with methods for finding neural network
architectures suitable for learning particular problems. We use an
evolutionary algorithm with four different genetic encoding methods to
search for the suitable architectures. We train the neural network
weights with a separate neural learning algorithm. We use eight
different learning problems for benchmarking the encoding
methods. Four of the problems are artificial (XOR, Encoder and two
function approximation problems), three are real-world classification
problems from the Proben1 benchmarking problem set, and one is a
bankruptcy classification problem studied earlier in one of our
projects. Our evaluation criteria are classification accuracy and
efficiency for using only the relevant variables. The classification
results are compared also to those for network architectures found by
a systematic search.
File: Evolutionary Design of Neural Networks (1018 kB) - the thesis in
compressed PostScript (.ps.gz, may not always be available online)
-- Marko Grönroos, magi AT iki PISTE fi (http://www.iki.fi/~magi/)
-- Evolutionary algorithm researchers do it with the fittest individuals --
|
