A STRUCTURAL LEARNING OF NEURAL-NETWORK CLASSIFIERS USING PCA NETWORKS AND SPECIES GENETIC ALGORITHMS

We present experimental results for a structural learning method of fe ed-forward neural-network classifiers using Principal Component Analys is (PCA) network and Species Genetic Algorithm (SGA). PCA network is u sed as a means for reducing the number of input units. SGA, a modified GA, is employed for selecting the proper number of hidden units and o ptimizing the connection links. Experimental results show that the pro posed method is a useful tool for choosing an appropriate architecture for high dimensions.