Performance Measures on Jumping Gene - Multiobjective Optimization Methodology: A Jumping Gene Approach - page 99

0.04
0.03
0.02
0.01
0

0.8
0.6
0.4
0.2
0

35

1

30

35

1 0.8

30 25

0.8

25 0.6

20

0.6

20

15

0.4

15 0.4

10

10

0.2

0.2

5 5

0 0

parameter index jumping rate parameter index jumping rate

Performance on convergence Performance on diversity

(a)

0.05
0.04
0.03
0.02
0.01
0

1
0.8
0.6
0.4
0.2
0

1

25 1 25

0.8 0.8

20 20

0.6 0.6

15 15

0.4 0.4

10 10

0.2 0.2

5 5

parameter index
Performance on convergence

0

jumping rate parameter index

0

jumping rate

Performance on diversity

(b)

Figure 5.1 (See COlOr iNSerT.)
3D.plots.for.average.values.of.convergence.and.diversity.metrics.against.different.parameter.
indexes.and.jumping.rates.for.different.test.functions..(a).Unconstrained.test.function:.FON;.
(b).unconstrained.test.function:.POL;.(c).constrained.test.function:.SRIN;.(d).constrained.test.
function:.TAN..(From.Chan,.T..M.,.Man,.K..F.,.Kwong,.S.,.Tang,.K..S.,.A.jumping.gene.paradigm.
for. evolutionary. multiobjective. optimization,. IEEE Transactions on Evolutionary Computation ,.
12(2),.143-159,.2008.)

The.results.for.different.algorithms.under.different.performance.metrics.
are.presented.in.the.following.discussion.

5.7.2.1 Mean and Standard Deviation of Generational
Distance for Evaluating Convergence

The.means.and.standard.deviations.of.the.generational.distance.convergence.
metric.for.the.unconstrained.test.functions.appear.in . Table 5.4 .. From.the.table,.
the.JG.found.the.most.convergent.nondominated.solutions.for.the.test.func-
tions.ZIT3.and.ZIT4.(in.binary.coding).and.ZIT1-ZIT4.and.ZIT6.(in.real-number.
coding)..Also,.the.JG.was.the.first.runner-up.for.the.remaining.test.functions.
The.means.and.standard.deviations.of.the.generational.distance.for.the.con-
strained.test.functions.are.given.in . Table 5.5 .. From.the.table,.although.the.JG.