Applications of Artificial Intelligence Techniques to Enhance Sustainability of Industry 4.0: Design of an Artificial Neural Network Model as Dynamic Behavior Optimizer of Robotic Arms

Table 7

Neural network training results for Stanford arm.

Number of neurons	25	20	15
Number of iterations	11	21	28
Number of training samples	700	700	700
Number of validation samples	150	150	150
Number of testing samples	150	150	150
Training mean square error	0.59	0.61	0.68
Validation mean square error	0.63	0.18	0.23
Test mean square error	0.39	0.55	0.25
Total mean square error	0.54	0.44	0.46
Training regression	0.7555	0.8012	0.8318
Validation regression	0.7178	0.1051	0.4116
Test regression	0.6074	0.5734	0.4426
Total regression	0.7141	0.3640	0.6491