ec.gp.breed

Class MutateDemotePipeline

java.lang.Object

ec.BreedingSource

ec.BreedingPipeline

ec.gp.GPBreedingPipeline

ec.gp.breed.MutateDemotePipeline

All Implemented Interfaces:

Prototype, Setup, SteadyStateBSourceForm, RandomChoiceChooserD, Serializable, Cloneable

public class MutateDemotePipeline
extends GPBreedingPipeline

MutateDemotePipeline works very similarly to the DemoteNode algorithm described in Kumar Chellapilla, "A Preliminary Investigation into Evolving Modular Programs without Subtree Crossover", GP98, and is also similar to the "insertion" operator found in Una-May O'Reilly's thesis, "An Analysis of Genetic Programming".

MutateDemotePipeline tries picks a random tree, then picks randomly from all the demotable nodes in the tree, and demotes one. If its chosen tree has no demotable nodes, or demoting its chosen demotable node would make the tree too deep, it repeats the choose-tree-then-choose-node process. If after tries times it has failed to find a valid tree and demotable node, it gives up and simply copies the individual.

"Demotion" means to take a node n and insert a new node m between n and n's parent. n becomes a child of m; the place where it becomes a child is determined at random from all the type-compatible slots of m. The other child slots of m are filled with randomly-generated terminals. Chellapilla's version of the algorithm always places n in child slot 0 of m. Because this would be unneccessarily restrictive on strong typing, MutateDemotePipeline instead picks the slot at random from all available valid choices.

A "Demotable" node means a node which is capable of demotion given the existing function set. In general to demote a node foo, there must exist in the function set a nonterminal whose return type is type-compatible with the child slot foo holds in its parent; this nonterminal must also have a child slot which is type-compatible with foo's return type.

This method is very expensive in searching nodes for "demotability". However, if the number of types is 1 (the GP run is typeless) then the type-constraint-checking code is bypassed and the method runs a little faster.

Typical Number of Individuals Produced Per produce(...) call
...as many as the source produces

Number of Sources
1

Parameters

base.tries int >= 1	(number of times to try finding valid pairs of nodes)
base.maxdepth int >= 1	(maximum valid depth of a mutated tree)
base.tree.0 0 < int < (num trees in individuals), if exists	(tree chosen for mutation; if parameter doesn't exist, tree is picked at random)

Default Base
gp.breed.mutate-demote

Version:

1.0

Author:

Sean Luke

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
private GPNode	demotableNode
(package private) int	maxDepth The maximum depth of a mutated tree
static int	NUM_SOURCES
(package private) int	numTries The number of times the pipeline tries to build a valid mutated tree before it gives up and just passes on the original
static String	P_MAXDEPTH
static String	P_MUTATEDEMOTE
static String	P_NUM_TRIES
(package private) int	tree Is our tree fixed? If not, this is -1

Fields inherited from class ec.gp.GPBreedingPipeline

P_NODESELECTOR, P_TREE, TREE_UNFIXED

Fields inherited from class ec.BreedingPipeline

DYNAMIC_SOURCES, likelihood, mybase, P_LIKELIHOOD, P_NUMSOURCES, P_SOURCE, sources, V_SAME, V_STUB

Fields inherited from class ec.BreedingSource

NO_PROBABILITY, P_PROB, probability

Constructor Summary

Constructors

Constructor and Description
MutateDemotePipeline()

Method Summary

Modifier and Type	Method and Description
private void	_demoteSomething(GPNode node, EvolutionState state, int thread, GPFunctionSet set)
private void	_demoteSomethingTypeless(GPNode node, EvolutionState state, int thread, GPFunctionSet set)
private int	_numDemotableNodes(GPInitializer initializer, GPNode root, int soFar, GPFunctionSet set)
private int	_pickDemotableNode(GPInitializer initializer, GPNode root, int num, GPFunctionSet set)
Parameter	defaultBase() Returns the default base for this prototype.
private boolean	demotable(GPInitializer initializer, GPNode node, GPFunctionSet set)
private void	demoteSomething(GPNode node, EvolutionState state, int thread, GPFunctionSet set)
private int	numDemotableNodes(GPInitializer initializer, GPNode root, int soFar, GPFunctionSet set)
int	numSources() Returns the number of sources to this pipeline.
private int	pickDemotableNode(GPInitializer initializer, GPNode root, int num, GPFunctionSet set)
int	produce(int min, int max, int subpopulation, ArrayList<Individual> inds, EvolutionState state, int thread, HashMap<String,Object> misc) Produces n individuals from the given subpopulation and puts them into inds[start...start+n-1], where n = Min(Max(q,min),max), where q is the "typical" number of individuals the BreedingSource produces in one shot, and returns n.
void	setup(EvolutionState state, Parameter base) Sets up the BreedingPipeline.
private boolean	verifyPoint(GPNode inner1) Returns true if inner1's depth + atdepth +1 is within the depth bounds

Methods inherited from class ec.gp.GPBreedingPipeline

produces

Methods inherited from class ec.BreedingPipeline

clone, fillStubs, finishProducing, individualReplaced, maxChildProduction, minChildProduction, preparePipeline, prepareToProduce, sourcesAreProperForm, typicalIndsProduced

Methods inherited from class ec.BreedingSource

getProbability, pickRandom, setProbability, setupProbabilities

Methods inherited from class java.lang.Object

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

Field Detail

P_MUTATEDEMOTE

public static final String P_MUTATEDEMOTE

See Also:

Constant Field Values

P_NUM_TRIES

public static final String P_NUM_TRIES

See Also:

Constant Field Values

P_MAXDEPTH

public static final String P_MAXDEPTH

See Also:

Constant Field Values

NUM_SOURCES

public static final int NUM_SOURCES

See Also:

Constant Field Values

numTries

int numTries

The number of times the pipeline tries to build a valid mutated tree before it gives up and just passes on the original

maxDepth

int maxDepth

The maximum depth of a mutated tree

tree

int tree

Is our tree fixed? If not, this is -1

demotableNode

private GPNode demotableNode

Constructor Detail

MutateDemotePipeline

public MutateDemotePipeline()

Method Detail

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(...).

numSources

public int numSources()

Description copied from class: BreedingPipeline

Returns the number of sources to this pipeline. Called during BreedingPipeline's setup. Be sure to return a value > 0, or DYNAMIC_SOURCES which indicates that setup should check the parameter file for the parameter "num-sources" to make its determination.

Specified by:

numSources in class BreedingPipeline

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 BreedingPipeline

See Also:

Prototype.setup(EvolutionState,Parameter)

demotable

private boolean demotable(GPInitializer initializer,
                          GPNode node,
                          GPFunctionSet set)

demoteSomething

private void demoteSomething(GPNode node,
                             EvolutionState state,
                             int thread,
                             GPFunctionSet set)

_demoteSomething

private void _demoteSomething(GPNode node,
                              EvolutionState state,
                              int thread,
                              GPFunctionSet set)

_demoteSomethingTypeless

private void _demoteSomethingTypeless(GPNode node,
                                      EvolutionState state,
                                      int thread,
                                      GPFunctionSet set)

numDemotableNodes

private int numDemotableNodes(GPInitializer initializer,
                              GPNode root,
                              int soFar,
                              GPFunctionSet set)

_numDemotableNodes

private int _numDemotableNodes(GPInitializer initializer,
                               GPNode root,
                               int soFar,
                               GPFunctionSet set)

pickDemotableNode

private int pickDemotableNode(GPInitializer initializer,
                              GPNode root,
                              int num,
                              GPFunctionSet set)

_pickDemotableNode

private int _pickDemotableNode(GPInitializer initializer,
                               GPNode root,
                               int num,
                               GPFunctionSet set)

verifyPoint

private boolean verifyPoint(GPNode inner1)

Returns true if inner1's depth + atdepth +1 is within the depth bounds

produce

public int produce(int min,
                   int max,
                   int subpopulation,
                   ArrayList<Individual> inds,
                   EvolutionState state,
                   int thread,
                   HashMap<String,Object> misc)

Description copied from class: BreedingSource

Produces n individuals from the given subpopulation and puts them into inds[start...start+n-1], where n = Min(Max(q,min),max), where q is the "typical" number of individuals the BreedingSource produces in one shot, and returns n. max must be >= min, and min must be >= 1. For example, crossover might typically produce two individuals, tournament selection might typically produce a single individual, etc.

Specified by:

produce in class BreedingSource