Package ec.steadystate

Class SteadyStateEvolutionState

java.lang.Object

ec.EvolutionState

ec.steadystate.SteadyStateEvolutionState

All Implemented Interfaces:

Setup, Singleton, Serializable

public class SteadyStateEvolutionState
extends EvolutionState

This subclass of EvolutionState implements basic Steady-State Evolution and (in distributed form) Asynchronous Evolution. The procedure is as follows. We begin with an empty Population and one by one create new Indivdiuals and send them off to be evaluated. In basic Steady-State Evolution the individuals are immediately evaluated and we wait for them; but in Asynchronous Evolution the individuals are evaluated for however long it takes and we don't wait for them to finish. When individuals return they are added to the Population until it is full. No duplicate individuals are allowed.

At this point the system switches to its "steady state": individuals are bred from the population one by one, and sent off to be evaluated. Once again, in basic Steady-State Evolution the individuals are immediately evaluated and we wait for them; but in Asynchronous Evolution the individuals are evaluated for however long it takes and we don't wait for them to finish. When an individual returns, we mark an individual in the Population for death, then replace it with the new returning individual. Note that during the steady-state, Asynchronous Evolution could be still sending back some "new" individuals created during the initialization phase, not "bred" individuals.

The determination of how an individual is marked for death is done by the SteadyStateBreeder.

SteadyStateEvolutionState will run either for some N "generations" or for some M evaluations of individuals. A "generation" is defined as a Population's worth of evaluations. If you do not specify the number of evaluations (the M), then SteadyStateEvolutionState will use the standard generations parameter defined in EvolutionState.

Parameters

evaluations int >= 1	(maximal number of evaluations to run.)
steady.replacement-probability 0.0 <= double <= 1.0 (default is 1.0)	(probability that an incoming individual will unilaterally replace the individual marked for death, as opposed to replacing it only if the incoming individual is superior in fitness)

See Also:

• Serialized Form

Field Summary

Fields

Modifier and Type	Field	Description
boolean	emptyAtGenerationBoundary	If true, the population will be emptied after each "generation," so no replacement or breeding occurrs.
protected boolean	firstTime	First time calling evolve
boolean	generationBoundary	Did we just start a new generation?
int	generationSize	how big is a generation? Set to the size of subpopulation 0 of the initial population.
static final String	P_EMPTY_AT_GEN
static final String	P_REPLACEMENT_PROBABILITY
double	replacementProbability	When a new individual arrives, with what probability should it directly replace the existing "marked for death" individual, as opposed to only replacing it if it's superior?

Fields inherited from class ec.EvolutionState

breeder, breedthreads, C_STARTED_FRESH, C_STARTED_FROM_CHECKPOINT, checkpoint, checkpointDirectory, checkpointModulo, checkpointPrefix, data, evalthreads, evaluations, evaluator, exchanger, finisher, generation, initializer, innovationNumber, job, numEvaluations, numGenerations, output, P_BREEDER, P_CHECKPOINT, P_CHECKPOINTDIRECTORY, P_CHECKPOINTMODULO, P_CHECKPOINTPREFIX, P_EVALUATIONS, P_EVALUATOR, P_EXCHANGER, P_FINISHER, P_GENERATIONS, P_INITIALIZER, P_INNOVATIONNUMBER, P_QUITONRUNCOMPLETE, P_STATISTICS, parameters, population, quitOnRunComplete, R_FAILURE, R_NOTDONE, R_SUCCESS, random, randomSeedOffset, runtimeArguments, statistics, UNDEFINED

Constructor Summary

Constructors

Constructor	Description
SteadyStateEvolutionState()

Method Summary

Modifier and Type	Method	Description
int	evolve()
void	finish(int result)
void	finishEvaluationStatistics()
void	setup(EvolutionState state, Parameter base)	Unlike for other setup() methods, ignore the base; it will always be null.
void	startFresh()

Methods inherited from class ec.EvolutionState

incrementEvaluations, resetFromCheckpoint, run, startFromCheckpoint

Methods inherited from class java.lang.Object

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

Field Details

P_REPLACEMENT_PROBABILITY

public static final String P_REPLACEMENT_PROBABILITY

See Also:

• Constant Field Values

P_EMPTY_AT_GEN

public static final String P_EMPTY_AT_GEN

See Also:

• Constant Field Values

generationBoundary

public boolean generationBoundary

Did we just start a new generation?

generationSize

public int generationSize

how big is a generation? Set to the size of subpopulation 0 of the initial population.

replacementProbability

public double replacementProbability

When a new individual arrives, with what probability should it directly replace the existing "marked for death" individual, as opposed to only replacing it if it's superior?

emptyAtGenerationBoundary

public boolean emptyAtGenerationBoundary

If true, the population will be emptied after each "generation," so no replacement or breeding occurrs. This is used, for example, by Ant Colony algorithms, which have no notion of breeding.

firstTime

protected boolean firstTime

First time calling evolve

Constructor Details

SteadyStateEvolutionState

public SteadyStateEvolutionState()

Method Details

setup

public void setup(EvolutionState state,
 Parameter base)

Description copied from class: EvolutionState

Unlike for other setup() methods, ignore the base; it will always be null.

Specified by:

setup in interface Setup

Overrides:

setup in class EvolutionState

See Also:

• Prototype.setup(EvolutionState,Parameter)

startFresh

public void startFresh()

Overrides:

startFresh in class EvolutionState

evolve

public int evolve()

Overrides:

evolve in class EvolutionState

finishEvaluationStatistics

public void finishEvaluationStatistics()

finish

public void finish(int result)

Overrides:

finish in class EvolutionState