Information processing of an RBF network in Neural Networks free pdf
An RBF network receives the input through the unweight connections. Then the input vector is delivered through a norm so that the result is a scalar. This scalar is processed by a radial basis function.
The output values of the totally different neurons n of the RBF layer or maybe of the different Gaussianbells are added within the third layer basically, in terms of the whole input space, Gaussian bells are added here.