source: trunk/sources/HeuristicLab.Algorithms.ParameterlessPopulationPyramid/3.3/HillClimber.cs @ 14185

Last change on this file since 14185 was 14185, checked in by swagner, 3 years ago

#2526: Updated year of copyrights in license headers

File size: 4.5 KB
Line 
1#region License Information
2/* HeuristicLab
3 * Copyright (C) 2002-2016 Heuristic and Evolutionary Algorithms Laboratory (HEAL)
4 * and the BEACON Center for the Study of Evolution in Action.
5 *
6 * This file is part of HeuristicLab.
7 *
8 * HeuristicLab is free software: you can redistribute it and/or modify
9 * it under the terms of the GNU General Public License as published by
10 * the Free Software Foundation, either version 3 of the License, or
11 * (at your option) any later version.
12 *
13 * HeuristicLab is distributed in the hope that it will be useful,
14 * but WITHOUT ANY WARRANTY; without even the implied warranty of
15 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
16 * GNU General Public License for more details.
17 *
18 * You should have received a copy of the GNU General Public License
19 * along with HeuristicLab. If not, see <http://www.gnu.org/licenses/>.
20 */
21#endregion
22
23using System;
24using System.Collections.Generic;
25using System.Linq;
26using System.Threading;
27using HeuristicLab.Common;
28using HeuristicLab.Core;
29using HeuristicLab.Data;
30using HeuristicLab.Encodings.BinaryVectorEncoding;
31using HeuristicLab.Optimization;
32using HeuristicLab.Parameters;
33using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
34using HeuristicLab.Problems.Binary;
35using HeuristicLab.Random;
36
37
38namespace HeuristicLab.Algorithms.ParameterlessPopulationPyramid {
39  // This code is based off the publication
40  // B. W. Goldman and W. F. Punch, "Parameter-less Population Pyramid," GECCO, pp. 785–792, 2014
41  // and the original source code in C++11 available from: https://github.com/brianwgoldman/Parameter-less_Population_Pyramid
42  [Item("Hill Climber (HC)", "Binary Hill Climber.")]
43  [StorableClass]
44  [Creatable(CreatableAttribute.Categories.SingleSolutionAlgorithms, Priority = 150)]
45  public class HillClimber : BasicAlgorithm {
46    [Storable]
47    private IRandom random;
48
49    private const string IterationsParameterName = "Iterations";
50
51    public override Type ProblemType {
52      get { return typeof(BinaryProblem); }
53    }
54    public new BinaryProblem Problem {
55      get { return (BinaryProblem)base.Problem; }
56      set { base.Problem = value; }
57    }
58
59    public IFixedValueParameter<IntValue> IterationsParameter {
60      get { return (IFixedValueParameter<IntValue>)Parameters[IterationsParameterName]; }
61    }
62
63    public int Iterations {
64      get { return IterationsParameter.Value.Value; }
65      set { IterationsParameter.Value.Value = value; }
66    }
67
68    [StorableConstructor]
69    protected HillClimber(bool deserializing) : base(deserializing) { }
70    protected HillClimber(HillClimber original, Cloner cloner)
71      : base(original, cloner) {
72    }
73    public override IDeepCloneable Clone(Cloner cloner) {
74      return new HillClimber(this, cloner);
75    }
76
77    public HillClimber()
78      : base() {
79      random = new MersenneTwister();
80      Parameters.Add(new FixedValueParameter<IntValue>(IterationsParameterName, "", new IntValue(100)));
81    }
82    protected override void Run(CancellationToken cancellationToken) {
83      var BestQuality = new DoubleValue(double.NaN);
84      Results.Add(new Result("Best quality", BestQuality));
85      for (int iteration = 0; iteration < Iterations; iteration++) {
86        var solution = new BinaryVector(Problem.Length);
87        for (int i = 0; i < solution.Length; i++) {
88          solution[i] = random.Next(2) == 1;
89        }
90
91        var fitness = Problem.Evaluate(solution, random);
92
93        fitness = ImproveToLocalOptimum(Problem, solution, fitness, random);
94        if (double.IsNaN(BestQuality.Value) || Problem.IsBetter(fitness, BestQuality.Value)) {
95          BestQuality.Value = fitness;
96        }
97      }
98    }
99    // In the GECCO paper, Section 2.1
100    public static double ImproveToLocalOptimum(BinaryProblem problem, BinaryVector solution, double fitness, IRandom rand) {
101      var tried = new HashSet<int>();
102      do {
103        var options = Enumerable.Range(0, solution.Length).Shuffle(rand);
104        foreach (var option in options) {
105          if (tried.Contains(option)) continue;
106          solution[option] = !solution[option];
107          double newFitness = problem.Evaluate(solution, rand);
108          if (problem.IsBetter(newFitness, fitness)) {
109            fitness = newFitness;
110            tried.Clear();
111          } else {
112            solution[option] = !solution[option];
113          }
114          tried.Add(option);
115        }
116      } while (tried.Count != solution.Length);
117      return fitness;
118    }
119  }
120}
Note: See TracBrowser for help on using the repository browser.