### Table 1* Back Propagation Genetic Algorithm

1995

"... In PAGE 11: ... In each case network architecture is that of Figure 1. The parameters used in the training of the network are given in Table1 . Two measures of fit, namely the sum of squares deviation (Ssq.... In PAGE 17: ... Table1 . The parameter settings to run the neural network using both the back propagation and the genetic algorithms to run all three problems are given.... ..."

### Table 19: Results of the Back-Propagation algorithm on the two spiral problem. CCR: correct classi cation rate.

1995

### Table 18: Results of the Error Back-Propagation algorithm on the two spiral prob- lem. CCR: correct classi cation rate.

1995

### Table 4. Variables optimized using neural networks trained with AC algorithm and back- propagation algorithm

### Table 1 Summary of Properties of the CHIR and the Common Error Back-propagation Training Algorithms for Multi-layer Perceptrons

### Table 2: Performance and programming time for back-propagation.

"... In PAGE 7: ... To get an idea of the di erences, the back-propagation was programmed three times: in C and Assembly language for MUSIC (data-parallel code) and in standard C for a single-processor computer. The results are summarized in Table2 . Notice that the programming times are estimated and only the kernel part of the algorithm is counted in Table 2.... In PAGE 7: ... The results are summarized in Table 2. Notice that the programming times are estimated and only the kernel part of the algorithm is counted in Table2 . The shell part of all versions is programmed... ..."

### Table 8: The performance of the back propagation algorithm with the basic features compared to its performance with the features generated by ficus. The numbers in parentheses are con dence intervals with p =0:95.

2002

Cited by 16

### Table 4 Estimation error by identified hybrid models (Back-propagation method)

"... In PAGE 9: ... The performance values were calculated based on two independent experiments, which were different from the 20 training experiments. TABLE 4 and TABLE 5 Table4 shows some comparisons of the numerical results of back-propagation training algorithm without interpolation, with linear interpolation and with cubic spline interpolation. In the first case, there was not interpolation, so the training data only contained the measurement data (with 1 or 2 hour sample time).... ..."