Expected
class ExpectedImprovement(var xi: Double = BayesianOptimizationSolver.defaultXi) : AcquisitionFunctionIfc(source)
Expected Improvement (EI) for minimization. With incumbent f+, predictive mean μ, and predictive standard deviation σ, the improvement is I = f+ - μ - xi, and EI = I·Φ(I/σ) + σ·φ(I/σ) for σ > 0, else max(I, 0). This is the default acquisition.
Parameters
xi
an exploration margin subtracted from the improvement (larger encourages exploration). Defaults to BayesianOptimizationSolver.defaultXi.
Functions
Link copied to clipboard
open override fun value(prediction: SurrogateModelIfc.Prediction, incumbent: Double, bo: BayesianOptimizationSolver): Double
The acquisition value (larger is better) for a candidate.