Package org.moeaframework.problem.DTLZ

Class DTLZ5

java.lang.Object
org.moeaframework.problem.AbstractProblem
org.moeaframework.problem.DTLZ.DTLZ
org.moeaframework.problem.DTLZ.DTLZ5
All Implemented Interfaces:
AutoCloseable, Problem, AnalyticalProblem
public class DTLZ5 extends DTLZ
The DTLZ5 test problem.

Note: The original DTLZ5 and DTLZ6 problems have a known defect where they produce additional solutions than expected. See Deb and Saxena (2006) for more details.

References:

  1. Deb, K. and Saxena, D. K. (2006). Searching for Pareto-optimal solutions through dimensionality reduction for certain large-dimensional multi-objective optimization problems. In The 2006 IEEE Congress on Evolutionary Computation, pages 3353–3360.

  • Constructor Details

    • DTLZ5

      public DTLZ5(int numberOfObjectives)
      Constructs a DTLZ5 test problem with the specified number of objectives. This is equivalent to calling new DTLZ5(numberOfObjectives+9, numberOfObjectives).
      Parameters:
      numberOfObjectives - the number of objectives for this problem

    • DTLZ5

      public DTLZ5(int numberOfVariables, int numberOfObjectives)
      Constructs a DTLZ5 test problem with the specified number of variables and objectives.
      Parameters:
      numberOfVariables - the number of variables for this problem
      numberOfObjectives - the number of objectives for this problem

  • Method Details

    • evaluate

      public void evaluate(Solution solution)
      Description copied from interface: Problem
      Evaluates the solution, updating the solution's objectives in place. Algorithms must explicitly call this method when appropriate to evaluate new solutions or reevaluate modified solutions.
      Parameters:
      solution - the solution to be evaluated

    • generate

      public Solution generate()
      Description copied from interface: AnalyticalProblem
      Returns a randomly-generated solution using the analytical solution to this problem. Note however that discontinuous Pareto surfaces may result in some solutions generated by this method being dominated by other generated solutions. It is therefore recommended using a NondominatedPopulation to remove dominated solutions prior to using the generated reference set.

      The generated solutions should be spread uniformly across the entire Pareto frontier; however, this is a suggestion and is not a requirement of this interface.

      Returns:
      a randomly-generated Pareto optimal solution to this problem