Table 6.

Parameter grid and intervals used in Bayesian Optimisation procedure

	Interval	Description
Eta	[0.01, 1]	Learning rate (shrinkage applied in weights calculation)
Gamma	[0, 100]	Minimum loss reduction to split a node in tree
Max_depth	[1, 9]	Maximum depth of each tree in training process
Subsample	[0.5, 1]	Number of features used to train a tree
Lambda	[1, 100]	$L_2$ regularization term using in training
Alpha	[0, 100]	$L_1$ regularization term using in training
n_Estimators	[10, 200]	Total number of trees