org.cicirello.replication.ieeetevc2016.FDC Maven / Gradle / Ivy

Go to download
/*
 * Example programs for JavaPermutationTools library.
 * Copyright (C) 2015-2016, 2018, 2021-2023 Vincent A. Cicirello
 *
 * This program is free software: you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation, either version 3 of the License, or
 * (at your option) any later version.
 *
 * This program is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU General Public License for more details.
 *
 * You should have received a copy of the GNU General Public License
 * along with this program.  If not, see .
 */

package org.cicirello.replication.ieeetevc2016;

import org.cicirello.examples.jpt.ExamplesShared;
import org.cicirello.math.stats.Statistics;
import org.cicirello.permutations.*;
import org.cicirello.permutations.distance.*;

/**
 * This program computes the fitness distance correlations for the "Permutation in a Haystack"
 * fitness landscapes found in Table II of the paper:

 * V.A. Cicirello, "The
 * Permutation in a Haystack Problem and the Calculus of Search Landscapes," IEEE Transactions
 * on Evolutionary Computation, 20(3):434-446, June 2016.
 *
 * NOTE: This is not the original code from this paper. Source code has been heavily refactored
 * since then, but results should correspond.
 *
 * @author Vincent A. Cicirello, https://www.cicirello.org/
 */
public class FDC {

  /**
   * Runs the replication program.
   *
   * @param args Ignored, because there are no command line arguments.
   */
  public static void main(String[] args) {
    ExamplesShared.printCopyrightAndLicense();

    // list of distance metrics used
    PermutationDistanceMeasurer[] d = new PermutationDistanceMeasurer[26];
    d[0] = new ExactMatchDistance();
    d[1] = new InterchangeDistance();
    d[2] = new DeviationDistance();
    d[3] = new AcyclicEdgeDistance();
    d[4] = new CyclicEdgeDistance();
    d[5] = new RTypeDistance();
    d[6] = new CyclicRTypeDistance();
    d[7] = new KendallTauDistance();
    d[8] = new ReinsertionDistance();

    // warning: this next is slow: brute force breadth first computation of reversal distance for
    // all 10! permutations of length 10
    d[9] = new ReversalDistance(10);

    d[10] = new ScrambleDistance();
    d[11] = new ReversalIndependentDistance(d[7]);
    d[12] = new ReversalIndependentDistance(d[1]);
    d[13] = new ReversalIndependentDistance(d[8]);
    d[14] = new ReversalIndependentDistance(d[9]);
    d[15] = new ReversalIndependentDistance(d[10]);
    d[16] = new CyclicReversalIndependentDistance(d[7]);
    d[17] = new CyclicReversalIndependentDistance(d[1]);
    d[18] = new CyclicReversalIndependentDistance(d[8]);
    d[19] = new CyclicReversalIndependentDistance(d[9]);
    d[20] = new CyclicReversalIndependentDistance(d[10]);
    d[21] = new CyclicIndependentDistance(d[7]);
    d[22] = new CyclicIndependentDistance(d[1]);
    d[23] = new CyclicIndependentDistance(d[8]);
    d[24] = new CyclicIndependentDistance(d[9]);
    d[25] = new CyclicIndependentDistance(d[10]);

    // table element [i][j] must be correlation coefficient of distance d[i] and distance
    // d[distanceIndices[i][j]]
    int[][] distanceIndices = {
      {7, 1, 8, 9, 10}, // 0
      {7, 1, 8, 9, 10}, // 1
      {7, 1, 8, 9, 10}, // 2
      {11, 12, 13, 14, 15}, // 3 need reversal independence
      {16, 17, 18, 19, 20}, // 4 need reversal and cyclic independence
      {7, 1, 8, 9, 10}, // 5
      {21, 22, 23, 24, 25}, // 6 need cyclic independence
      {7, 1, 8, 9, 10}, // 7
      {7, 1, 8, 9, 10} // 8
    };

    Permutation p = new Permutation(10, 0);

    final int FACT = 3628800;

    int[][] data = new int[d.length][FACT];

    // warning: following iterates over all permutations of length 10
    // Compute distance to reference permutation from all permutations of length 10 for all
    // distances in d
    int i = 0;
    for (Permutation q : p) {
      for (int j = 0; j < d.length; j++) {
        data[j][i] = d[j].distance(p, q);
      }
      i++;
    }

    System.out.println(
        "Table of fitness distance correlations for several \"Permutation in a Haystack\" landscapes and distances.");
    System.out.println(
        "Essentially generates the data in Table 2 of \"The Permutation in a Haystack Problem and the Calculus of Search Landscapes\"");
    System.out.println(
        "by Vincent A. Cicirello, from IEEE Transactions on Evolutionary Computation, 20(3): 434-446, June 2016.");
    System.out.println(
        "NOTE: This is not the original code.  Source code has been heavily refactored since, but results should correspond.");

    System.out.println();
    System.out.println(
        "Haystack Landscape \t    AdjSwap\tInterchange\tReinsertion\t   Reversal\t   Scramble");
    for (i = 0; i < distanceIndices.length; i++) {
      System.out.printf("%19s", d[i].getClass().getSimpleName());
      for (int j = 0; j < distanceIndices[i].length; j++) {
        int k = distanceIndices[i][j];
        System.out.printf("\t%11.3f", Statistics.correlation(data[i], data[k]));
      }
      System.out.println();
    }
  }
}