### Table 4. Classification performance of RBF networks

"... In PAGE 8: ... The RBF training algorithm was stopped after adding 50, 100, 200 and 400 neurons. The RBF network performance ( Table4 ) was best for spread constant 20 and 100 allocated neurons (72.... ..."

### Table 3: Classi cation using RBF networks

### Table 1: Size and performance of the RBF networks used.

### Table 1 Parameters set of a RBF network

"... In PAGE 10: ... Only those hypotheses that are found to be useful survive through generations. An experimental justification for the use of grow- ing is as follows: In all experiments performed so far, the number of basis function was limited to 15 basis functions (see the number of regions parameter from Table1 , which also indicates the maximum number of basis function). This low limit was used because if GAs start with a higher limit (e.... ..."

### Table 2. Character classification rates using the RBF Neural Network

"... In PAGE 3: ... Two tables are presented which give the highest printed and handwritten character recognition rates. Table 1 presents results for the Backpropagation neural network and Table2 presents results for the RBF neural network. Table 1.... ..."

### Table 4. Comparative results (heart) - MLP, RBF and CASCOR networks

"... In PAGE 7: ... This consisted of considering the networks that presented the `best apos; results in the previous experiments for repeating each network simulation a number of 10 runs and calculating statistics on the distributions of the results obtained (the best run rate, the mean of the rates and the standard deviation). Table 3 and Table4 summarize all the results obtained in the three different sets for the card problem and the heart problem, respectively. These tables include the best results reported with the MLP networks [15] and the best results obtained with the experimentation carried out with the RBF networks.... ..."

### Table 4. The prediction accuracy of RBF networks. The RBF network constructed with 50 genes selected by the P-metric value shows the best performance.

"... In PAGE 12: ... So, we ran the active RAN algorithm on the second and the third set of genes ten times respectively. Table4 shows the result of each experiment. The RBF network with 50 gene expression levels classifies all training samples correctly and its test error is 1.... ..."

### Table 1. Pseudo-Code for Kalman Training of RBF Networks

1995

Cited by 5

### TABLE II RESULTS USING THE RBF NEURAL NETWORK CLASSIFIER Verification Rate [%]

2006

Cited by 1