net.sourceforge.cilib.problem.coevolution
Class CompetitiveCoevolutionGameLearningOptimizationProblem

java.lang.Object
  net.sourceforge.cilib.problem.OptimisationProblemAdapter
      net.sourceforge.cilib.problem.coevolution.PerformanceEvaluationOptimizationProblem
          net.sourceforge.cilib.problem.coevolution.GameLearningOptimizationProblem
              net.sourceforge.cilib.problem.coevolution.CompetitiveCoevolutionGameLearningOptimizationProblem

All Implemented Interfaces:

Serializable, CoevolutionOptimisationProblem, OptimisationProblem, Problem, Cloneable

public class CompetitiveCoevolutionGameLearningOptimizationProblem
extends GameLearningOptimizationProblem
implements CoevolutionOptimisationProblem

This class is used by CoevolutionAlgorithm to optimize a Game.

Author:

leo

See Also:

Serialized Form

Field Summary

Fields inherited from class net.sourceforge.cilib.problem.coevolution.GameLearningOptimizationProblem

game

Fields inherited from class net.sourceforge.cilib.problem.coevolution.PerformanceEvaluationOptimizationProblem

fitnessCalculation, numberOfEvaluations

Fields inherited from class net.sourceforge.cilib.problem.OptimisationProblemAdapter

dataSetBuilder, fitnessEvaluations

Constructor Summary

CompetitiveCoevolutionGameLearningOptimizationProblem()

CompetitiveCoevolutionGameLearningOptimizationProblem(CompetitiveCoevolutionGameLearningOptimizationProblem copy)
Copy constructor

Method Summary

protected Fitness	calculateFitness(Type solution) Determine the Fitness of the current Problem instance based on the provided solution.
Fitness	evaluateEntity(int populationID, int evaluationRound, Blackboard<Enum<?>,Type> entityData) This method evaluates an entity with regards to a number of groups of entities depending on the instance of the problem
int	getAmountSubPopulations() this will correlate with the amount of groups of entites an individual required for evaluation
DomainRegistry	getSubPopulationDomain(int populationNo) This method is used by the coevolution algorithm to assign an optimization problem instance to each sub population this optimization problem needs to define the domain for that specific sub population
void	initializeEntities(PopulationBasedAlgorithm pba, int populationID) This method adds any problem specific entries to the all the populations entites's blackboards.

Methods inherited from class net.sourceforge.cilib.problem.coevolution.GameLearningOptimizationProblem

getBehaviouralDomain, getClone, getDomain, getGame, playGame, setGame

Methods inherited from class net.sourceforge.cilib.problem.coevolution.PerformanceEvaluationOptimizationProblem

getNumberOfEvaluations, setFitnessCalculation, setNumberOfEvaluations

Methods inherited from class net.sourceforge.cilib.problem.OptimisationProblemAdapter

accept, changeEnvironment, getChangeStrategy, getDataSetBuilder, getFitness, getFitnessEvaluations, setChangeStrategy, setDataSetBuilder

Methods inherited from class java.lang.Object

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

Constructor Detail

CompetitiveCoevolutionGameLearningOptimizationProblem

public CompetitiveCoevolutionGameLearningOptimizationProblem()

CompetitiveCoevolutionGameLearningOptimizationProblem

public CompetitiveCoevolutionGameLearningOptimizationProblem(CompetitiveCoevolutionGameLearningOptimizationProblem copy)

Copy constructor

Parameters:

copy -

Method Detail

evaluateEntity

public Fitness evaluateEntity(int populationID,
                              int evaluationRound,
                              Blackboard<Enum<?>,Type> entityData)

This method evaluates an entity with regards to a number of groups of entities depending on the instance of the problem

Specified by:

evaluateEntity in interface CoevolutionOptimisationProblem

entityData - the blackboard of the Entity being evaluated. This contains the solution vector, scoreboard and competitor list

Returns:

the entitie's fitness

calculateFitness

protected Fitness calculateFitness(Type solution)

Determine the Fitness of the current Problem instance based on the provided solution.

Overrides:

calculateFitness in class GameLearningOptimizationProblem

Parameters:

solution - The Type representing the candidate solution.

Returns:

The Fitness of the solution in the current Problem.

See Also:

OptimisationProblemAdapter.getFitness(Type, boolean)

initializeEntities

public void initializeEntities(PopulationBasedAlgorithm pba,
                               int populationID)

This method adds any problem specific entries to the all the populations entites's blackboards.

Specified by:

initializeEntities in interface CoevolutionOptimisationProblem

Parameters:

pba - the sub population to initialize

populationID - the populationID of the sub population

getAmountSubPopulations

public int getAmountSubPopulations()

this will correlate with the amount of groups of entites an individual required for evaluation

Specified by:

getAmountSubPopulations in interface CoevolutionOptimisationProblem

Returns:

the amount of sub population for this problem

getSubPopulationDomain

public DomainRegistry getSubPopulationDomain(int populationNo)

This method is used by the coevolution algorithm to assign an optimization problem instance to each sub population this optimization problem needs to define the domain for that specific sub population

Specified by:

getSubPopulationDomain in interface CoevolutionOptimisationProblem

Parameters:

populationNo - the population number for the sub population for which this problem is generated

Returns:

the optimization problem for the sub population