Kernel Parameter Optimization for Kriging Based on Structural Risk Minimization Principle

Table 2

Comparison of the generalization ability between different test functions with uniform distribution sampling.

Test functions				RMSE
	CON-Kriging	SRM-Kriging	MLE-Kriging	CON-Kriging	SRM-Kriging	MLE-Kriging
Fun1	Failure	0.999 39	0.672 59	0.372 85	0.005 82	0.134 61
Fun2	1.000 00	1.000 00	1.000 00	0.000 00	0.000 00	0.000 00
Fun3	0.988 30	0.991 93	0.999 20	1.197 99	0.994 75	0.313 37
Fun4	0.950 40	0.980 54	0.979 91	103.012 78	64.513 05	65.463 16
Fun5	0.894 21	0.993 22	0.986 45	326.976 85	82.749 71	117.016 48
Fun6	0.999 93	0.999 99	0.999 98	0.694 53	0.263 94	0.341 66
Fun7	0.723 31	0.969 00	0.899 10	11.309 11	3.783 29	6.824 92
Fun8	Failure	0.931 03	0.105 14	168.556 75	42.903 68	154.449 27
Fun9	Failure	0.755 16	0.179 30	1 010.540 9	499.434 93	913.770 52