Cover of Swarm Intelligence Book
Andries P Engelbrecht Fundamentals of Computational Swarm Intelligence

Evolutionary Computation Algorithms

The following classes are available (all discussions assume CIlib version 0.5, and therefor assumes the same directory structure as used in CIlib):

  • Genetic Algorithm [ Show ]
  • Evolutionary Programming [ Show ]
  • Evolutionary Strategy [ Hide ]

    Find below the XML specification for a Evolutionary Strategy (section 9.4).

    <simulator>
      <algorithms>
        <algorithmid="es"class="ec.ea.ES">
          <populationSize>
            30
          </populationSize>
          <prototypeIndividualclass="ec.ea.Individual"/>
          <lambdaclass="controlparameterupdatestrategies.ConstantUpdateStrategy"parameter="200.0"/>
          <parentSelectorclass="ec.selectionoperators.RandomSelection"/>
          <crossProbabilityclass="controlparameterupdatestrategies.ConstantUpdateStrategy"parameter="1.0"/>
          <crossoverclass="ec.crossoveroperators.DiscreteRecombination"parents="2"/>
          <mutationProbabilityclass="controlparameterupdatestrategies.ConstantUpdateStrategy"parameter="1.0"/>
          <mutatorclass="ec.mutationoperators.ASMutationDecorator">
            <componentclass="ec.mutationoperators.ESDeviationMutation">
              <randomSampleclass="math.RandomNumber"/>
            </component>
            <problemidref="spherical"/>
          </mutator>
          <parentsAsWell>
            false
          </parentsAsWell>
          <nextGenerationSurvivalSelectorclass="ec.selectionoperators.Elitism"/>
          <addStoppingConditionclass="stoppingcondition.MaximumIterations"iterations="1000"/>
        </algorithm>
      </algorithms>
      <problems>
        <problemid="spherical"class="problem.FunctionMinimisationProblem">
          <functionclass="functions.continuous.Spherical"domain="R(-5.12,5.12)^30"behavioralDomain="R(0,1)"/>
        </problem>
        <problemid="rosenbrock"class="problem.FunctionMinimisationProblem">
          <functionclass="functions.continuous.Rosenbrock"domain="R(-2.048,2.048)^30"behavioralDomain="R(0,1)"/>
        </problem>
        <problemid="griewank"class="problem.FunctionMinimisationProblem">
          <functionclass="functions.continuous.Griewank"domain="R(-600,600)^30"behavioralDomain="R(0,1)"/>
        </problem>
        <problemid="hyperellipsoid"class="problem.FunctionMinimisationProblem">
          <functionclass="functions.continuous.HyperEllipsoid"domain="R(-1,1)^30"behavioralDomain="R(0,1)"/>
        </problem>
        <problemid="colville"class="problem.FunctionMinimisationProblem">
          <functionclass="functions.continuous.Colville"domain="R(-10,10)^4"behavioralDomain="R(0,1)"/>
        </problem>
        <problemid="ackley"class="problem.FunctionMinimisationProblem">
          <functionclass="functions.continuous.Ackley"domain="R(-30,30)^30"behavioralDomain="R(0,1)"/>
        </problem>
      </problems>
      <measurementsid="measurements"class="simulator.MeasurementSuite"samples="1"resolution="100">
        <addMeasurementclass="measurement.single.FunctionOptimisationError"/>
        <addMeasurementclass="measurement.single.Solution"/>
      </measurements>
      <simulations>
        <simulation>
          <algorithmidref="es"/>
          <problemidref="spherical"/>
          <measurementsidref="measurements"file="data/es-spherical.txt"/>
        </simulation>
        <simulation>
          <algorithmidref="es"/>
          <problemidref="rosenbrock"/>
          <measurementsidref="measurements"file="data/es-rosenbrock.txt"/>
        </simulation>
        <simulation>
          <algorithmidref="es"/>
          <problemidref="griewank"/>
          <measurementsidref="measurements"file="data/es-griewank.txt"/>
        </simulation>
        <simulation>
          <algorithmidref="es"/>
          <problemidref="hyperellipsoid"/>
          <measurementsidref="measurements"file="data/es-hyperellipsoid.txt"/>
        </simulation>
        <simulation>
          <algorithmidref="es"/>
          <problemidref="ackley"/>
          <measurementsidref="measurements"file="data/es-ackley.txt"/>
        </simulation>
        <simulation>
          <algorithmidref="es"/>
          <problemidref="colville"/>
          <measurementsidref="measurements"file="data/es-colville.txt"/>
        </simulation>
      </simulations>
    </simulator>

    Click here to download this file.

  • Differential Evolution [ Show ]