Lower
class LowerConfidenceBound(beta: Double = BayesianOptimizationSolver.defaultBeta) : AcquisitionFunctionIfc(source)
Lower Confidence Bound (LCB) for minimization. Minimizing μ - β·σ is equivalent to maximizing β·σ - μ, which is the value returned here. Larger beta favors exploration.
Parameters
beta
the exploration weight on the predictive standard deviation. Must be >= 0. Defaults to BayesianOptimizationSolver.defaultBeta.
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.