Table 8 Results of the BELBFM model vs. known regression models on the test dataset (test 6).

BELBFM

See Table 3

Neural network

Preset: narrow neural network | number of fully connected layers: 1 | first layer size: 10 | activation: ReLU | iteration limit: 1000 | Regularization (lambda) = 0 | standardize data: yes

Neural network

Preset: medium neural network | number of fully connected layers: 1 | first layer size: 25 | activation: ReLU | iteration limit: 1000 | regularization (lambda) = 0 | standardize data: yes

Neural network

Preset: wide neural network | number of fully connected layers: 1 | first layer size: 100 | activation: ReLU | iteration limit: 1000 | regularization (lambda) = 0 | standardize data: yes

Neural network

Preset: bilayered neural network | number of fully connected layers: 2 | first layer size: 10 | second layer size: 10 | activation: ReLU | iteration limit: 1000 | regularization (lambda) = 0 | standardize data: yes

Neural network

Preset: trilayered neural network | number of fully connected layers: 3 | first layer size: 10 | second layer size: 10 | third layer size: 10 | activation: ReLU | iteration limit: 1000 | regularization (lambda) = 0 | standardize data: yes

Ensemble

Preset: boosted trees | minimum leaf size: 8 | number of learners: 30 | learning rate: 0.1

Ensemble

Preset: bagged trees | minimum leaf size: 8 | number of learners: 30

Kernel

Preset: SVM kernel | learner: SVM | number of expansion dimensions: auto | regularization strength (lambda): auto | kernel scale: auto | epsilon: auto | standardize data: yes | iteration limit: 1000

Kernel

Preset: least squares regression kernel | learner: least squares kernel | number of expansion dimensions: auto | regularization strength (lambda): auto | kernel scale: auto | epsilon: auto | standardize data: yes | iteration limit: 1000

Gaussian process regression

Preset: rational quadratic GPR | basic function: constant | kernel function: rational quadratic | use isotropic kernel: yes | kernel scale: automatic | signal standard deviation: automatic | sigma: automatic | standardize data: yes | optimize numeric parameters: yes

Gaussian process regression

Preset: squared exponential GPR | basic function: constant | kernel function: squared exponential | use isotropic kernel: yes | kernel scale: automatic | signal standard deviation: automatic | sigma: automatic | standardize data: yes | optimize numeric parameters: yes

Gaussian process regression

Preset: 5/2 GPR | basic function: constant | kernel function: matem 5/2 | use isotropic kernel: yes | kernel scale: automatic | signal standard deviation: automatic | sigma: automatic | standardize data: yes | optimize numeric parameters: yes

Gaussian process regression

Preset: Exponential GPR | basic function: constant | kernel function: exponential | use isotropic kernel: yes | kernel scale: automatic | signal standard deviation: automatic | sigma: automatic | standardize data: yes | optimize numeric parameters: yes

Efficient linear

Preset: efficient linear least square | linear: least square | regularization: auto | regularization strength (lambda): auto | relative coefficient tolerance (beta tolerance) 0.0001

Efficient linear

Preset: efficient linear SVM | linear: SVM | regularization: auto | regularization strength (lambda): auto | relative coefficient tolerance (beta tolerance) 0.0001 | epsilon: auto

SVM

Preset: linear SVM | kernel function: linear | kernel scale: automatic | box constraint: automatic | epsilon: auto | standardize data: yes

SVM

Preset: quadratic SVM | kernel function: quadratic | kernel scale: automatic | box constraint: automatic | epsilon: auto | standardize data: yes

SVM

Preset: cubic SVM | kernel function: cubic | kernel scale: automatic | box constraint: automatic | epsilon: auto | standardize data: yes

SVM

Preset: fine gaussian SVM | kernel function: gaussian | kernel scale: 1 | box constraint: automatic | epsilon: auto | standardize data: yes

SVM

Preset: medium gaussian SVM | kernel function: gaussian | kernel scale: 4.1 | box constraint: automatic | epsilon: auto | standardize data: yes

SVM

Preset: coarse gaussian SVM | kernel function: gaussian | kernel scale: 16 | box constraint: automatic | epsilon: auto | standardize data: yes

Tree

Preset: fine tree | minimum leaf size: 4 | surrogate decision splits: off

Tree

Preset: medium tree | minimum leaf size: 12 | surrogate decision splits: off

Tree

Preset: coarse tree | minimum leaf size: 36 | surrogate decision splits: off

Linear regression

Preset: linear | terms: linear | robust optional: off

Linear regression

Preset: interactions linear | terms: interactions | robust optional: off

Linear regression

Preset: robust linear | terms: linear | robust optional: on

Stepwise linear regression

Preset: stepwise linear | initial terms: linear | upper bound on terms: interactions | maximum number of steps: 1000