Information Technology Reference
In-Depth Information
best individual of each generation (simple elitism), works very well, allow-
ing a very efficient search through the fitness landscape.
Finally, another important difference between natural systems and com-
puter systems is that in computer systems it is possible to measure rigorously
the fitness as we know exactly what lies ahead and what we want and, there-
fore, only individuals more or less fit to do a predetermined job are selected.
Consequently, it is fundamental the way we analyze the task at hand and
choose the conditions (selection environment or fitness cases) under which
individuals breed and are selected because, for once, we are probably going
to get what we asked for.
1.4 Genetic Algorithms
Genetic Algorithms were invented by John Holland in the 1960s and they
apply biological evolution theory to computer systems (Holland 1975). And
like all evolutionary computer systems, GAs are an oversimplification of
biological evolution. In this case, solutions to a problem are usually en-
coded in fixed length strings of 0's and 1's (chromosomes), and populations
of such strings (individuals or candidate solutions) are manipulated in or-
der to evolve a good solution to a particular problem. From generation to
generation individuals are reproduced with modification and selected ac-
cording to fitness. Modification in the original genetic algorithm was in-
troduced by the search operators of mutation, crossover, and inversion, but
more recent applications started favoring mutation and crossover, drop-
ping inversion in the process.
It is worth pointing out that GAs' individuals consist of naked chromo-
somes or, in other words, GAs' individuals are simple replicators. And like
all simple replicators, the chromosomes of GAs work both as genotype and
phenotype. This means that they are simultaneously the objects of selection
and the guardians of the genetic information that must be replicated and
passed on with modification to the next generation. Consequently, whatever
is done in the genome will affect fitness and selection. To make this impor-
tant feature of GAs clearer, compare this situation with the current state of
nature where individuals are selected by virtue of the properties of their bod-
ies alone: only the body of the individual and the abilities it can perform are
important to the selection process; the state of its genome is irrelevant.
