Package org.moeaframework.core.indicator
Class R2Indicator
java.lang.Object
org.moeaframework.core.indicator.NormalizedIndicator
org.moeaframework.core.indicator.RIndicator
org.moeaframework.core.indicator.R2Indicator
- All Implemented Interfaces:
Indicator
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:
- Hansen, M. P. and A. Jaszkiewicz (1998). Evaluating the Quality of Approximations to the Non-dominated Set. IMM Technical Report IMM-REP-1998-7.
-
Nested Class Summary
Nested classes/interfaces inherited from class org.moeaframework.core.indicator.RIndicator
RIndicator.ChebyshevUtility, RIndicator.LinearWeightedSumUtility, RIndicator.UtilityFunction
-
Field Summary
Fields inherited from class org.moeaframework.core.indicator.RIndicator
utilityFunction, weights
Fields inherited from class org.moeaframework.core.indicator.NormalizedIndicator
problem
-
Constructor Summary
ConstructorDescriptionR2Indicator
(Problem problem, int subdivisions, NondominatedPopulation referenceSet) Constructs a new R2 indicator using the Chebyshev utility function.R2Indicator
(Problem problem, int subdivisions, NondominatedPopulation referenceSet, Normalizer normalizer, RIndicator.UtilityFunction utilityFunction) Constructs a new R2 indicator using the specified utility function. -
Method Summary
Modifier and TypeMethodDescriptiondouble
evaluate
(NondominatedPopulation population) Returns the value of this unary quality indicator given the specified non-dominated population.Methods inherited from class org.moeaframework.core.indicator.RIndicator
expectedUtility, getDefaultSubdivisions
Methods inherited from class org.moeaframework.core.indicator.NormalizedIndicator
getNormalizedReferenceSet, normalize
-
Constructor Details
-
R2Indicator
Constructs a new R2 indicator using the Chebyshev utility function. The default normalization procedure, as specified byDefaultNormalizer
, is used.- Parameters:
problem
- the problemsubdivisions
- the number of subdivisions along each objectivereferenceSet
- 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 problemsubdivisions
- the number of subdivisions along each objectivereferenceSet
- the reference setnormalizer
- the user-provided normalizer, ornull
if the default is usedutilityFunction
- the utility function
-
-
Method Details
-
evaluate
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
-