|
The development of Artificial Neural Networks (ANNs) is traditionally a slow process in which human
experts are needed to experiment on differen tarchitectural procedures until they find the one that
presents the correct results that solvea specific problem. This work describes a new technique that uses
Genetic Programming(GP) in order to automatically develop simple ANNs, with a low number of
neurons and connections. Experiments havebeencarriedoutinordertomeasurethebehaviorofthe
systemandalsotocomparetheresultsobtainedusingotherANNgenerationandtrainingmethodswith
evolutionarycomputation(EC)tools.Theobtainedresultsare,intheworstcase,atleastcomparableto
existingtechniquesand,inmanycases,substantiallybetter.Asexplainedherein,thesystemhasother
importantfeaturessuchasvariablediscrimination,whichprovidesnewinformationontheproblemsto
be solved.
doi:10.1016/j.neucom.2010.05.010
|
.: SABIA :. Sistemas Adaptativos y Bioinspirados en Inteligencia Artificial