Hyperparameter Tuning of Machine Learning Algorithms Using Response Surface Methodology: A Case Study of ANN, SVM, and DBN

Table 18

ANOVA on MAE of testing set by hyperparameter tuning methods: GS and RSM.

Source	Df	Adj SS	Adj MS	F-value	value
Fold	9	0.1723	0.0191	10.47	<0.001
ML	2	0.0367	0.0183	10.03	<0.001
Tuning method	1	0.0044	0.0044	2.41	0.127
Error	47	0.085969	0.001829
Total	59	0.299374