Package ec.select

Class BoltzmannSelection

java.lang.Object
ec.BreedingSource
ec.SelectionMethod
ec.select.FitProportionateSelection
ec.select.BoltzmannSelection
All Implemented Interfaces:
Prototype, Setup, RandomChoiceChooserD, Serializable, Cloneable
public class BoltzmannSelection extends FitProportionateSelection
Similar to FitProportionateSelection, but with a Simulated Annealing style twist. BoltzmannSelection picks individuals of a population in proportion to an adjusted version of their fitnesses instead of their actual fitnesses as returned by fitness(). The adjusted fitness is calculated by e^(fitness/current_temperature) where current_temperature is a temperature value that decreases by a constant cooling rate as generations of evolution pass. The current_temperature is calculated by starting-temperature - (cooling-rate * the_current_generation_number). When the temperature dips below 1.0, annealing ceases and BoltzmannSelection reverts to normal FitProportionateSelection behavior.

Like FitProportionateSelection this is not appropriate for steady-state evolution. If you're not familiar with the relative advantages of selection methods and just want a good one, use TournamentSelection instead. Not appropriate for multiobjective fitnesses.

Note: Fitnesses must be non-negative. 0 is assumed to be the worst fitness.

Typical Number of Individuals Produced Per produce(...) call
Always 1.

Parameters

base.starting-temperature
double = some large number (defaults to 1.0)		 (the starting temperature for our simulated annealing style adjusted fitness proportions)
base.cooling-rate
double = some smaller number (defaults to 0.0 which causes BoltzmannSelection to behave just as FitProportionateSelection would)		 (how slow, or fast, do you want to cool the annealing fitness proportions?)

Default Base
select.boltzmann

See Also:

  • Field Details

  • Constructor Details

    • BoltzmannSelection

      public BoltzmannSelection()

  • Method Details

    • defaultBase

      public Parameter defaultBase()
      Description copied from interface: Prototype
      Returns the default base for this prototype. This should generally be implemented by building off of the static base() method on the DefaultsForm object for the prototype's package. This should be callable during setup(...).
      Specified by:
      defaultBase in interface Prototype
      Overrides:
      defaultBase in class FitProportionateSelection

    • setup

      public void setup(EvolutionState state, Parameter base)
      Description copied from class: BreedingSource
      Sets up the BreedingPipeline. You can use state.output.error here because the top-level caller promises to call exitIfErrors() after calling setup. Note that probability might get modified again by an external source if it doesn't normalize right.

      The most common modification is to normalize it with some other set of probabilities, then set all of them up in increasing summation; this allows the use of the fast static BreedingSource-picking utility method, BreedingSource.pickRandom(...). In order to use this method, for example, if four breeding source probabilities are {0.3, 0.2, 0.1, 0.4}, then they should get normalized and summed by the outside owners as: {0.3, 0.5, 0.6, 1.0}.

      Specified by:
      setup in interface Prototype
      Specified by:
      setup in interface Setup
      Overrides:
      setup in class BreedingSource
      See Also:

    • prepareToProduce

      public void prepareToProduce(EvolutionState s, int subpopulation, int thread)
      Description copied from class: SelectionMethod
      A default version of prepareToProduce which does nothing.
      Overrides:
      prepareToProduce in class FitProportionateSelection