Class R2Indicator

All Implemented Interfaces:
Indicator

public class R2Indicator extends RIndicator
Computes the R2 indicator. The R2 indicator is the expected utility evaluated across a set of uniformly-weighted utility functions. R2 is weakly compatible with the outperformance relation for any set of utility functions. Values are normalized between [-1, 1] with -1 preferred.

References:

  1. Hansen, M. P. and A. Jaszkiewicz (1998). Evaluating the Quality of Approximations to the Non-dominated Set. IMM Technical Report IMM-REP-1998-7.
  • Constructor Details

    • R2Indicator

      public R2Indicator(Problem problem, int subdivisions, NondominatedPopulation referenceSet)
      Constructs a new R2 indicator using the Chebyshev utility function. The default normalization procedure, as specified by DefaultNormalizer, is used.
      Parameters:
      problem - the problem
      subdivisions - the number of subdivisions along each objective
      referenceSet - the reference set
    • R2Indicator

      public R2Indicator(Problem problem, int subdivisions, NondominatedPopulation referenceSet, Normalizer normalizer, RIndicator.UtilityFunction utilityFunction)
      Constructs a new R2 indicator using the specified utility function.
      Parameters:
      problem - the problem
      subdivisions - the number of subdivisions along each objective
      referenceSet - the reference set
      normalizer - the user-provided normalizer, or null if the default is used
      utilityFunction - the utility function
  • Method Details

    • evaluate

      public double evaluate(NondominatedPopulation population)
      Description copied from interface: Indicator
      Returns the value of this unary quality indicator given the specified non-dominated population.
      Parameters:
      population - the non-dominated population to be evaluated
      Returns:
      the value of this quality indicator given the specified non-dominated population