📄 bulkfitnessoffsetremover.java
字号:
/*
* This file is part of JGAP.
*
* JGAP offers a dual license model containing the LGPL as well as the MPL.
*
* For licencing information please see the file license.txt included with JGAP
* or have a look at the top of class org.jgap.Chromosome which representatively
* includes the JGAP license policy applicable for any file delivered with JGAP.
*/
package org.jgap.impl;
import java.util.*;
import org.jgap.*;
/**
* <p>
* Takes away the fitness offset of the population to evolve.
* The fitness function values of the population of {@link org.jgap.Chromosome}
* instances will start from a minimum of 1 afterwards.
* </p>
* <p>
* The removal of an offset in the fitness values of a population strengthens the
* "survival of the fittest" effect of a selector that performs selection
* upon fitness values. A high offset in the fitness values of a population
* lowers the relative difference between the fitness values of the Chromosomes in
* a population.
* </p>
* <h3>Example of applicability</h3>
* <p>
* You are optimizing a black box with <i>n</i> parameters that are mapped
* to {@link org.jgap.Chromosome} instances each having <i>n</i> {@link org.jgap.Gene}
* instances.<br>
* You want to minimize the answer time of the black box and provide
* a {@link org.jgap.FitnessFunction#evaluate(org.jgap.Chromosome)}
* that takes the genes out of the chromosome, put's it's {@link org.jgap.Gene#getAllele()}
* values to the parameters and measures the answer time of the black box
* (by invoking it's service to optimize). <br>
* The longer the time takes, the worse it's fitness is, so you have to
* invert the measured times to fitness values:
* <a name="bboptimizer"/>
* <pre>
* <font color="#0011EE">
* class BlackBoxOptimizer extends org.jgap.FitnessFunction{
* private BlackBox bbox;
* <font color="#999999">//Additional code: constructors</font>
* <font color="#999999">...</font>
* public double evaluate(org.jgap.Chromosome chromosome){
* double fitness = 0;
* <font color="#999999">// get the Gene[] & put the parameters into the box.
* ...
* </font>
* long duration = System.currentTimeMillis(); <font color="#999999">// You certainly will use an advanced StopWatch...</font>
* this.bbox.service(); <font color="#999999">// The black boxes service to optimize.</font>
* duration = System.currentTimeMillis()-duration;
* <font color="#999999">// transform the time into fitness value:</font>
* fitness = double.MAX_VALUE - (double)duration;
* return fitness;
* }
* }
* </font>
* </pre>
* </p>
* <p>
* <h4>We might get the following results (each row stands for a Chromosome, the table is a population):</h4>
* <table border="1">
* <tr align="left" valign="top">
* <th>
* duration
* </th>
* <th>
* fitness
* </th>
* <th>
* piece of fitness cake
* </th>
* </tr>
* <tr align="left" valign="top">
* <td>
* 2000
* </td>
* <td>
* 9218868437227403311
* </td>
* <td>
* 33.333333333333336949106088992532 %
* </td>
* </tr>
* <tr align="left" valign="top">
* <td>
* 3000
* </td>
* <td>
* 9218868437227402311
* </td>
* <td>
* 33.333333333333333333333333333333 %
* </td>
* </tr>
* <tr align="left" valign="top">
* <td>
* 4000
* </td>
* <td>
* 9218868437227401311
* </td>
* <td>
* 33.333333333333329717560577674135 %
* </td>
* </tr>
* </table>
* </p>
* <p>
* If any {@link org.jgap.NaturalSelector} performs selection based upon the fitness values, it
* will have to put those values in relation to each other. As a fact, the probability
* to select the Chromosome that contained the black box parameters that caused an answer time
* of 4000 ms is "equal" to the probability to select the Chromosome that caused a black box
* answer time to be 2000 ms!
* </p>
* <p>
* Of course one could work around that problem by replacing the <tt>Integer.MAX_VALUE</tt>
* transformation by a fixed maximum value the black box would need for the service.
* But what, if you have no guaranteed maximum answer time for the service of the black box ?
* Even if you have got one, it will be chosen sufficently high above the average answer
* time thus letting your fitness function return values with a high offset in the fitness.
* </p>
* <p>
* <h4>This is, what happens, if you use this instance for fitness evaluation:</h4>
* <table border="1">
* <tr align="left" valign="top">
* <th>
* duration
* </th>
* <th>
* fitness
* </th>
* <th>
* piece of fitness cake
* </th>
* </tr>
* <tr align="left" valign="top">
* <td>
* 2000
* </td>
* <td>
* 2001
* </td>
* <td>
* 66.63 %
* </td>
* </tr>
* <tr align="left" valign="top">
* <td>
* 3000
* </td>
* <td>
* 1001
* </td>
* <td>
* 33.33 %
* </td>
* </tr>
* <tr align="left" valign="top">
* <td>
* 4000
* </td>
* <td>
* 1
* </td>
* <td>
* 0.03 %
* </td>
* </tr>
* </table>
* </p>
* <h3>Example of usage</h3>
*
* <p>
* This example shows how to use this instance for cutting fitness offsets.
* It is the same example as used <a href="#bboptimizer">above</a>.
* <pre>
* <font color="#0011EE">
* class BlackBoxOptimizer extends org.jgap.FitnessFunction{
* <font color="#999999">// Additional code: constructors
* ...</font>
* public double evaluate(org.jgap.Chromosome chromosome){
* <font color="#999999">.... // As shown above.</font>
* }
*
* public void startOptimization(org.jgap.Configuration gaConf)throws InvalidConfigurationException{
* <font color="#999999">// The given Configuration may be preconfigured with
* // NaturalSelector & GeneticOperator instances,.
* // But should not contain a FitnessFunction or BulkFitnessFunction!</font>
* <b>gaConf.setBulkFitnessFunction(new BulkFitnessOffsetRemover(this));</b>
* <font color="#999999">// Why does it work? We implement FitnessFunction!
* // Still to do here:
* // - Create a sample chromosome according to your blackbox & set it to the configuration.
* // - Create a random inital Genotype.
* // - loop over a desired amount of generations invoking aGenotype.evolve()..</font>
* }
* }
* </font>
* </pre>
* </p>
* @author Achim Westermann
* @since 2.2
*
*/
public class BulkFitnessOffsetRemover
extends BulkFitnessFunction {
/** String containing the CVS revision. Read out via reflection!*/
private final static String CVS_REVISION = "$Revision: 1.3 $";
/*
* Replace this member by the Configuration as
* Replace this member by the Configuration as
* soon as Configuration allows bulk fitness function and
* fitness function to be stored both in it.
*/
private FitnessFunction ff;
⌨️ 快捷键说明
复制代码
Ctrl + C
搜索代码
Ctrl + F
全屏模式
F11
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -