(EX)
download PDF
with first 40 pages
from my latest eBook
if you need more operators
click here
Keywords
selection,
exploration,
exploitation,
exploration/exploitation balance,
competition for survival
Motivation
●
Assessing the effectivity of integrating the selection and
crossover processes.
Source text
●
Thierens D.,
Goldberg D.E. (1994),
Elitist Recombination: an integrated selection recombination GA, in
Proceedings of the First IEEE World Congress on Computational
Intelligence, pp. 508512
WEB:
http://intl.ieeexplore.ieee.org/xpl/abs_free.jsp?arNumber=349898
http://www.cs.uu.nl/groups/DSS/publications/
Read also
●
Vekaria
K.,
Clack C. (1998),
Selective Crossover in Genetic Algorithms: An Empirical Study, in
Proceedings of the fifth Conference on Parallel Problem Solving from
Nature, SpringerVerlag, pp. 438447
WEB:
http://citeseer.ifi.unizh.ch/vekaria98selective.html
http://citeseer.ist.psu.edu/vekaria98selective.html
●
Coli M.,
Genusso P.,
Palazzari P. (1996),
A New Crossover Operator for Genetic Algorithms, in IEEE
International Conference on Evolutionary Computation, pp. 201206
WEB:
http://citeseer.ifi.unizh.ch/coli96new.html
http://citeseer.ist.psu.edu/coli96new.html
See also
●
Best Schema Crossover
●
Selective Crossover2
●
Partially Randomized Crossover
●
Direct Design Variable Exchange Crossover
Algorithm
1.
for every generation of GA do
2.
randomly shuffle the entire population P(t)={A_{1}^{(t)},...,A_{Population_size}^{(t)}}
3.
for i = 1 to Population_size do
4.
create two vectors V_{1} and V_{2}:
V_{1}=A_{i}^{(t)}XA_{i+1}^{(t)}
V_{2}=A_{i}^{(t)}XA_{i+1}^{(t)}
5.
compute the fitness value of
V_{1}
and V_{2}
6.
insert best two vectors of {A_{i}^{(t)},A_{i+1}^{(t)},V_{1},V_{2}}
into the next
population P(t+1) as offspring
7.
i
= i + 2
8.
end do
where:
X – preferred crossover method
Comments
●
In the standard genetic algorithm, the selection process is always
preceded by the crossover process. In the EX method both of the
processes are integrated. During the first step the entire population is
randomly shuffled (row: 2). Then, from each successive pair of parental
vectors, two new vectors are created by crossover (row: 4). From a
“family” created in this way, two best vectors are singled out and
implemented as offspring to the next population (row: 6).
●
Application of elitist selection in the traditional way that is on
the level of the entire population may often be the reason for the
premature convergence of the algorithm. An EX elitist selection applied
on the “family” level (row: 6) eliminates this danger according to the
authors.
Experiment domains
●
bit counting function
●
fully deceptive trap function
Compared to
●
Uniform Crossover
