Free cookie consent management tool by TermsFeed Policy Generator
wiki:Manipulators

Version 1 (modified by mkofler, 14 years ago) (diff)

Added manipulator description page

Manipulators

Manipulators are HeuristicLab 3.3 operators that implement the IManipulator interface. Manipulators are specific for a particular solution encoding.


1. Manipulators for BinaryVectorEncoding

1.1 SinglePositionBitflipManipulator

Flips exactly one bit of a binary vector. It is implemented as described in (Michalewicz 1999).

1.2 SomePositionBitflipManipulator

Flips some bits of a binary vector, each position is flipped with a probability of pm. It is implemented as described in (Eiben and Smith 2003, p. 43).

2. Manipulators for IntegerVectorEncoding

2.1 UniformOnePositionManipulator

Uniformly distributed change of a single position of an integer vector. It is implemented as described in (Michalewicz 1999).

3. Manipulators for PermuationEncoding

3.1 InsertionManipulator

An operator which moves randomly one element to another position in the permutation (Insertion is a special case of Translocation). It is implemented as described in (Fogel 1988).

3.2 InversionManipulator

An operator which inverts a randomly chosen part of a permutation. It is implemented as described in (Eiben and Smith 2003).

3.3 MultiPermutationManipulator

Randomly selects and applies one of its manipulators every time it is called.

3.4 ScrambleManipulator

An operator which manipulates a permutation array by randomly scrambling the elements in a randomly chosen interval. It is implemented as described in (Syswerda 1991).

3.5 !Swap2Manipulator

An operator which manipulates a permutation array by swapping to randomly chosen elements. It is implemented as described in (Eiben and Smith 2003).

3.6 !Swap3Manipuator

An operator which manipulates a permutation array by swaping three randomly chosen elements. It is implemented such that first 3 positions are randomly chosen in the interval [0;N) with N = length of the permutation with all positions being distinct from each other. Then position 1 is put in place of position 3, position 2 is put in place of position 1 and position 3 is put in place of position 2.

3.7 TranslocationInversionManipulator

An operator which inverts a randomly chosen part of a permutation and inserts it at a random position. It is implemented as described in (Fogel 1993).

3.8 TranslocationManipulator

An operator which Manipulates a permutation array by moving a randomly chosen interval of elements to another (randomly chosen) position in the array. It is implemented as described in (Michalewicz 1992).

4. Manipulators for RealvectorEncoding

4.1 BreederGeneticAlgorithmManipulator

It is implemented as described by (Mühlenbein and Schlierkamp-Voosen 1993).

4.2 MichalewiczNonUniformAllPositionsManipulator

It is implemented as described in (Michalewicz 1999).

4.3 MichalewiczNonUniformOnePositionManipulator

It is implemented as described in (Michalewicz 1999).

4.4 MultiRealVectorManipulator

Randomly selects and applies one of its manipulators every time it is called.

4.5 NormalAllPositionsManipulator

This manipulation operator adds a value sigma_i * N(0,1) to the current value in each position i. The values for sigma_i are taken from the strategy vector, if there are less elements in the strategy vector than positions, then the strategy vector is cycled. It is implemented as described in (Beyer and Schwefel 2002).

4.6 PolynomialAllPositionManipulator

The polynomial manipulation is implemented as described in (Deb and Goyal 1996). In this operator it is performed on all positions of the real vector.

4.7 PolynomialOnePositionManipulator

The polynomial manipulation is implemented as described in (Deb and Goyal 1996). In this operator it is performed on a single randomly chosen position of the real vector.

4.8 UniformOnePositionManipulator

Changes a single position in the vector by sampling uniformly from the interval [Minimum_i, Maximum_i) in dimension i. It is implemented as described in (Michalewicz 1999).

5. Manipulators for SymbolicExpressionTreeEncoding

5.1 ArgumentCreater

Manipulates a symbolic expression by creating a new argument within one function-defining branch.

5.2 ArgumentDeleter

Manipulates a symbolic expression by deleting an argument from an existing function defining branch.

5.3 ArgumentDuplicater

Manipulates a symbolic expression by duplicating an existing argument node of a function-defining branch.

5.4 ChangeNodeTypeManipulation

Selects a random tree node and changes the symbol size.

5.5 FullTreeShaker

Manipulates all nodes that have local parameters.

5.6 MultiSymbolicExpressionTreeArchitectureManipulator

Randomly selects and applies one of its architecture manipulators every time it is called.

5.7 MultiSymbolicExpressionTreeManipulator

Randomly selects and applies one of its manipulators every time it is called.

5.8 OnePointShaker

Selects a random node with local parameters and manipulates the selected node.

5.9 SubroutineCreater

Manipulates a symbolic expression by adding one new function-defining branch containing a proportion of a preexisting branch and by creating a reference to the new branch.

5.10 SubroutineDeleter

Manipulates a symbolic expression by deleting a preexisting function-defining branch.

5.11 SubroutineDuplicater

Manipulates a symbolic expression by duplicating a preexisting function-defining branch.


References

  • Beyer, H.-G. and Schwefel, H.-P. 2002. Evolution Strategies - A Comprehensive Introduction Natural Computing, 1, pp. 3-52.
  • Deb, K. & Goyal, M. A. 1996. Combined Genetic Adaptive Search (GeneAS) for Engineering Design Computer Science and Informatics, 26, pp. 30-45.
  • Eiben, A.E. and Smith, J.E. 2003. Introduction to Evolutionary Computation. Natural Computing Series. Springer-Verlag Berlin Heidelberg.
  • Fogel, D.B. 1988. An Evolutionary Approach to the Traveling Salesman Problem, Biological Cybernetics, 60, pp. 139-144.
  • Michalewicz, Z. 1999. Genetic Algorithms + Data Structures = Evolution Programs. Third, Revised and Extended Edition, Spring-Verlag Berlin Heidelberg.
  • Mühlenbein, H. and Schlierkamp-Voosen, D. 1993. Predictive Models for the Breeder Genetic Algorithm - I. Continuous Parameter Optimization. Evolutionary Computation, 1(1), pp. 25-49.
  • Syswerda, G. 1991. Schedule Optimization Using Genetic Algorithms. In Davis, L. (Ed.) Handbook of Genetic Algorithms, Van Nostrand Reinhold, New York, pp 332-349.