Information Technology Reference
In-Depth Information
Table 6.8
Performance and settings used in the iris problem with multiple outputs.
Number of runs
100
Number of generations
20,000
Population size
30
Number of fitness cases
150
Function set of ADFs
+ - * /
Terminal set
d0 - d3
Head length
7
Gene length
15
Number of genes/ADFs
5
Function set of homeotic genes
+ - * /
Terminal set of homeotic genes
ADFs 0-4
Number of homeotic genes/cells
3
Head length of homeotic genes
7
Length of homeotic genes
15
Chromosome length
120
Mutation rate
0.044
Inversion rate
0.1
IS transposition rate
0.1
RIS transposition rate
0.1
One-point recombination rate
0.3
Two-point recombination rate
0.3
Gene recombination rate
0.3
Gene transposition rate
0.1
Mutation rate in homeotic genes
0.044
Inversion rate in homeotic genes
0.1
RIS transposition rate in homeotic genes
0.1
IS transposition rate in homeotic genes
0.1
Fitness function
Eq. (4.15)
Rounding threshold
0.5
Average best-of-run fitness
147.20
And as you can see in Tables 4.7 and 6.8, the multicellular system with
multiple outputs (MCS-MO, for short) is extremely efficient, significantly
surpassing the multigenic system with multiple outputs (average best-of-run
number of hits of 147.20 compared to 146.48). Furthermore, the best-of-
experiment model (4.20) designed with the GEP-MO algorithm, classified
correctly only 148 out of 150 fitness cases, whereas the best-of-experiment
 
Search WWH ::




Custom Search