org.evosuite.ga.metaheuristics.mosa

Class CrowdingDistance<T extends Chromosome>

java.lang.Object

org.evosuite.ga.metaheuristics.mosa.CrowdingDistance<T>

public class CrowdingDistance<T extends Chromosome>
extends Object

This class implements different variants of Crowding Distance for many-objective problems

Author:

Annibale Panichella

Constructor Summary

Constructors

Constructor and Description
CrowdingDistance()

Method Summary

Modifier and Type	Method and Description
protected void	crowdingDistanceAssignment(List<T> front, Set<FitnessFunction<T>> set) Method used to assign the 'traditional' Crowding Distance
void	fastEpsilonDominanceAssignment(List<T> front, Set<FitnessFunction<T>> set) This method implements a "fast" version of the variant of the crowding distance named "epsilon-dominance-assignment" proposed by K\"{o}ppen and Yoshida in : [1] Mario K\"{o}ppen and Kaori Yoshida, "Substitute Distance Assignments in NSGA-II for handling Many-objective Optimization Problems", Evolutionary Multi-Criterion Optimization, Volume 4403 of the series Lecture Notes in Computer Science pp 727-741.
void	subvectorDominanceAssignment(List<T> front, Set<FitnessFunction<T>> set) This method implements a variant of the crowding distance named "subvector-dominance-assignment" proposed by K\"{o}ppen and Yoshida in : [1] Mario K\"{o}ppen and Kaori Yoshida, "Substitute Distance Assignments in NSGA-II for handling Many-objective Optimization Problems", Evolutionary Multi-Criterion Optimization, Volume 4403 of the series Lecture Notes in Computer Science pp 727-741.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

CrowdingDistance

public CrowdingDistance()

Method Detail

crowdingDistanceAssignment

protected void crowdingDistanceAssignment(List<T> front,
                                          Set<FitnessFunction<T>> set)

Method used to assign the 'traditional' Crowding Distance

Parameters:

front - front of non-dominated solutions/tests

set - set of goals/targets (e.g., branches) to consider

subvectorDominanceAssignment

public void subvectorDominanceAssignment(List<T> front,
                                         Set<FitnessFunction<T>> set)

This method implements a variant of the crowding distance named "subvector-dominance-assignment" proposed by K\"{o}ppen and Yoshida in : [1] Mario K\"{o}ppen and Kaori Yoshida, "Substitute Distance Assignments in NSGA-II for handling Many-objective Optimization Problems", Evolutionary Multi-Criterion Optimization, Volume 4403 of the series Lecture Notes in Computer Science pp 727-741.

Parameters:

front - front of non-dominated solutions/tests

set - set of goals/targets (e.g., branches) to consider

fastEpsilonDominanceAssignment

public void fastEpsilonDominanceAssignment(List<T> front,
                                           Set<FitnessFunction<T>> set)

This method implements a "fast" version of the variant of the crowding distance named "epsilon-dominance-assignment" proposed by K\"{o}ppen and Yoshida in : [1] Mario K\"{o}ppen and Kaori Yoshida, "Substitute Distance Assignments in NSGA-II for handling Many-objective Optimization Problems", Evolutionary Multi-Criterion Optimization, Volume 4403 of the series Lecture Notes in Computer Science pp 727-741.

Parameters:

front - front of non-dominated solutions/tests

set - set of goals/targets (e.g., branches) to consider