|
|||||||||
PREV CLASS NEXT CLASS | FRAMES NO FRAMES | ||||||||
SUMMARY: INNER | FIELD | CONSTR | METHOD | DETAIL: FIELD | CONSTR | METHOD |
java.lang.Object | +--ec.BreedingSource | +--ec.SelectionMethod | +--ec.select.GreedyOverselection
GreedyOverselection is a SelectionMethod which implements Koza-style fitness-proportionate greedy overselection. Not appropriate for multiobjective fitnesses.
This selection method first divides individuals in a population into two groups: the "good" ("top") group, and the "bad" ("bottom") group. The best top percent of individuals in the population go into the good group. The rest go into the "bad" group. With a certain probability (determined by the gets setting), an individual will be picked out of the "good" group. Once we have determined which group the individual will be selected from, the individual is picked using fitness proportionate selection in that group, that is, the likelihood he is picked is proportionate to his fitness relative to the fitnesses of others in his group.
All this is expensive to set up and bring down, so it's 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.
Typical Number of Individuals Produced Per produce(...) call
Always 1.
Parameters
base.top 0.0 <= float <= 1.0 |
(the percentage of the population going into the "good" (top) group) |
base.gets 0.0 <= float <= 1.0 |
(the likelihood that an individual will be picked from the "good" group) |
Default Base
select.greedy
Field Summary | |
static int |
CHECKBOUNDARY
CheckBoundary is 8 |
float |
gets_n_percent
|
static java.lang.String |
P_GETS
|
static java.lang.String |
P_GREEDY
|
static java.lang.String |
P_TOP
|
float[] |
sortedFitOver
|
float[] |
sortedFitUnder
|
int[] |
sortedPop
Sorted population -- since I *have* to use an int-sized individual (short gives me only 16K), I might as well just have pointers to the population itself. |
float |
top_n_percent
|
Fields inherited from class ec.SelectionMethod |
INDS_PRODUCED |
Fields inherited from class ec.BreedingSource |
CHECKBOUNDARY,
DEFAULT_PRODUCED,
NO_PROBABILITY,
P_PROB,
probability,
UNUSED |
Constructor Summary | |
GreedyOverselection()
|
Method Summary | |
Parameter |
defaultBase()
Returns the default base for this prototype. |
void |
finishProducing(EvolutionState s,
int subpopulation,
int thread)
A default version of finishProducing, which does nothing. |
void |
prepareToProduce(EvolutionState s,
int subpopulation,
int thread)
A default version of prepareToProduce which does nothing. |
int |
produce(int subpopulation,
EvolutionState state,
int thread)
An alternative form of "produce" special to Selection Methods; selects an individual from the given subpopulation and returns its position in that subpopulation. |
void |
setup(EvolutionState state,
Parameter base)
Sets up the BreedingPipeline. |
Methods inherited from class ec.SelectionMethod |
preparePipeline,
produce,
produces,
typicalIndsProduced |
Methods inherited from class ec.BreedingSource |
getProbability,
pickRandom,
protoClone,
protoCloneSimple,
setProbability,
setupProbabilities |
Methods inherited from class java.lang.Object |
clone,
equals,
finalize,
getClass,
hashCode,
notify,
notifyAll,
toString,
wait,
wait,
wait |
Field Detail |
public static final int CHECKBOUNDARY
public float[] sortedFitOver
public float[] sortedFitUnder
public int[] sortedPop
public static final java.lang.String P_GREEDY
public static final java.lang.String P_TOP
public static final java.lang.String P_GETS
public float top_n_percent
public float gets_n_percent
Constructor Detail |
public GreedyOverselection()
Method Detail |
public Parameter defaultBase()
public void setup(EvolutionState state, Parameter base)
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}.
Prototype.setup(EvolutionState,Parameter)
public void prepareToProduce(EvolutionState s, int subpopulation, int thread)
public int produce(int subpopulation, EvolutionState state, int thread)
public void finishProducing(EvolutionState s, int subpopulation, int thread)
|
|||||||||
PREV CLASS NEXT CLASS | FRAMES NO FRAMES | ||||||||
SUMMARY: INNER | FIELD | CONSTR | METHOD | DETAIL: FIELD | CONSTR | METHOD |