Coevolutionary Feature Selection Strategy for RBFNN Classifier

This paper presents a new hybrid learning algorithm based on cooperative coevolutionary algorithm (Co-CEA) for designing the radial basis function neural network (RBFNN) classifiers with an inductive feature selection. The hidden layer design and the feature selection correspond to two subpopulations. Collaborations among the two subpopulations are formed to obtain complete solutions. Experimental results illustrate that the proposed algorithm is able to achieve both good RBFNN structures and significant feature sets.