Department of Computer

science University of
Peshawar
Introduction to Genetic Algorithms
(II)

Advanced Algorithms
AdvAlgo-Spr08

Genetic Algorithms: Example

Mechanics of GA are surprisingly
simple, not more complex than copying
strings and swapping partial strings.
Good results are obtained for many
practical problems consisting of
following three operators:
Reproduction
Crossover
Mutation
AdvAlgo-Spr08

Genetic Algorithms: Example

Reproduction (copying)
Strings copied according to their objective function values
f (biologists call it the fitness function).
Fitness is some measure of profit, utility or goodness to be
maximized.
Fittest strings have a higher probability of contributing
one or more offspring in the next generation.
In natural systems fitness determined by creatures ability
to survive predators, other obstacles to adulthood and
subsequent reproduction.
In our artificial environment objective function is the final
arbiter of the string-creatures
life or death.
AdvAlgo-Spr08

Genetic Algorithms: Example

Crossover
This proceeds in two steps:
1. Members of the copied strings are combined at
random.
2. Each pair of strings undergoes crossover (as
follows):
An integer position is selected uniformly at random
|1, l 1| where l is string length.
Two new strings are created by swapping all
characters between positions k + 1 and l
inclusively
AdvAlgo-Spr08

Genetic Algorithms: Example

Mutation
Although selective reproduction and crossover
generate bulk of next generation, but these
operations may miss-out potentially useful material
i.e. 1s and 0s at particular locations.
In GA mutation is the random (with small probability)
alteration of the value of a string position.

AdvAlgo-Spr08