Theoretical Analysis of Jumping Gene Operations - Multiobjective Optimization Methodology: A Jumping Gene Approach - page 85

70000

60000

50000

Copy-and-Paste
Cut-and-Paste
Crossover (Uniform)
Crossover (2-Point)
Mutation (0.5)
Mutation (0.05)
Crossover

40000

30000

20000

10000

0

0 1000 2000 3000 4000 5000

No. of Generations
(a)

70000

60000

50000

40000

Copy-and-Paste
Cut-and-Paste
Crossover
Mutation (0.5)
Mutation (0.05)
Random

30000

20000

10000

0

0 1000 2000 3000 4000 5000

No. of Generations
(b)

Figure 4.7 (See COlOr iNSerT.)
The.searching.abilities.of.different.operations.in.a.finite.population.with.initial.population.(a).
randomly.generated.and.(b).of.identical.chromosomes..(From.Tang,.K..S.,.Yin,.R..J.,.Kwong,.S.,.
Ng,. K.. T.,. Man,. K.. F.,. A. theoretical. development. and. analysis. of. jumping. gene. genetic. algo-
rithm,. IEEETransactionsonIndustrialInformatics ,.7(3),.2011,.408-418.)

Consider.a.particular.primary.schemata.competition.set.{000;.001;.010;.011;.
100;.101;.110;.111}.with.the.corresponding.average.fitness.{0;.4;.2;.6;.1;.5;.3; 7},.
respectively. . Figure 4.8a . shows.the.proportion.of.schemata.against.genera-
tions.based.on.the.conventional.GA..It.is.observed.that.the.best.schema.will.
dominate. the. population,. while. the. rest. survive. mainly. due. to. mutation.
operations..This.pattern.of.dynamics.highly.depends.on.the.presence.of.the.
fitness.function.and.the.selection.process.