-Unfortunately, in practical applications the number of components is
-usually unknown. Hence, the first coding scheme cannot be used
-systematically. Therefore, we provide a refinement of the second
-scheme: each output is learned by a different ANN. Table~\ref{tab3}
-presents the results for this approach. In any case, whatever the
-considered feedforward network topologies, the maximum epoch number,
-and the kind of iterations, the configuration success rate is slightly
-improved. Moreover, the strategies predictions rates reach almost
-12\%, whereas in Table~\ref{tab2} they never exceed 1.5\%. Despite of
-this improvement, a long term prediction of chaotic iterations still
-appear to be an open issue.
-
-\begin{table}
-\caption{Prediction success rates for split outputs.}
-\label{tab3}
-\centering
-\begin{tabular}{|c||c|c|c|}
-\hline
-\multicolumn{4}{|c|}{Networks topology: 3~inputs, 1~output, and one hidden layer} \\
-\hline
-\hline
-Epochs & 125 & 250 & 500 \\
-\hline
-\hline
-Chaotic & \multicolumn{3}{c|}{Output = Configuration} \\
-\hline
-10~neurons & 12.39\% & 14.06\% & 14.32\% \\
-25~neurons & 13.00\% & 14.28\% & 14.58\% \\
-40~neurons & 11.58\% & 13.47\% & 14.23\% \\
-\hline
-\hline
-Non chaotic & \multicolumn{3}{c|}{Output = Configuration} \\
-\cline{2-4}
-%Epochs & 125 & 250 & 500 \\
-\hline
-10~neurons & 76.01\% & 74.04\% & 78.16\% \\
-25~neurons & 76.60\% & 72.13\% & 75.96\% \\
-40~neurons & 76.34\% & 75.63\% & 77.50\% \\
-\hline
-\hline
-Chaotic/non chaotic & \multicolumn{3}{c|}{Output = Strategy} \\
-\cline{2-4}
-%Epochs & 125 & 250 & 500 \\
-\hline
-10~neurons & 0.76\% & 0.97\% & 1.21\% \\
-25~neurons & 1.09\% & 0.73\% & 1.79\% \\
-40~neurons & 0.90\% & 1.02\% & 2.15\% \\
-\hline
-\multicolumn{4}{c}{} \\
-\hline
-Epochs & 1000 & 2500 & 5000 \\
-\hline
-\hline
-Chaotic & \multicolumn{3}{c|}{Output = Configuration} \\
-\hline
-10~neurons & 14.51\% & 15.22\% & 15.22\% \\
-25~neurons & 16.95\% & 17.57\% & 18.46\% \\
-40~neurons & 17.73\% & 20.75\% & 22.62\% \\
-\hline
-\hline
-Non chaotic & \multicolumn{3}{c|}{Output = Configuration} \\
-\cline{2-4}
-%Epochs & 1000 & 2500 & 5000 \\
-\hline
-10~neurons & 78.98\% & 80.02\% & 79.97\% \\
-25~neurons & 79.19\% & 81.59\% & 81.53\% \\
-40~neurons & 79.64\% & 81.37\% & 81.37\% \\
-\hline
-\hline
-Chaotic/non chaotic & \multicolumn{3}{c|}{Output = Strategy} \\
-\cline{2-4}
-%Epochs & 1000 & 2500 & 5000 \\
-\hline
-10~neurons & 3.47\% & 9.98\% & 11.66\% \\
-25~neurons & 3.92\% & 8.63\% & 10.09\% \\
-40~neurons & 3.29\% & 7.19\% & 7.18\% \\
-\hline
-\end{tabular}
-\end{table}