Download Document

Advanced AI – Session 6
Genetic Algorithm
By: H.Nematzadeh
Objectives
• To understand the processes involved ie. GAs
Basic flows –operator and parameters (roles,
effects etc)
• To be able to apply GAs in solving optimisation
problems
Evolutionary computation
• we are products of evolution, and thus by
modelling the process of evolution, we might
expect to create intelligent behaviour.
Evolutionary computation simulates evolution
on a computer. The result of such a simulation
is a series of optimisation algorithms, usually
based on a simple set of rules. Optimisation
iteratively improves the quality of solutions
until an optimal, or at least feasible, solution is
found.
Nature like evolution is slow
• Evolution is a tortuously slow process from
the human perspective, but the simulation of
evolution on a computer does not take billions
of years!
Natural evolution
• Evolution can be seen as a process leading to the
maintenance of a population’s ability to survive
and reproduce in a specific environment. This
ability is called evolutionary fitness.
• Evolutionary fitness can also be viewed as a
measure of organism’s ability to anticipate
changes in its environment.
• The better an organism's fitness to the
environment, the better its chances to survive
Rabbits & Foxes
Encoding Vs Evaluation
Class of searches techniques
Evolutionary Process
Mice & Cats: an evolutionary problem
The mice & cat algorithm
General evolution process
GA Vs Real life
Basic GA
Basic GA
Another way of looking at this…
Flowchart of GA
Another way of looking at this…
GA Process
Example 1 (not included in the book)
burger and profit problem
Analysis
Fitness Evaluation
Selection
Crossover
Mutation
After 1st run
Example 2: optimization of a one
variable function
Steps in GA development
The entire universe of discourse
Operator parameters
Fitness function
The fitness functions and
chromosomes location
Selection using roulette wheel
• One of the most commonly used chromosome
selection techniques is the roulette wheel
selection (Goldberg, 1989; Davis, 1991). Figure
7.4 illustrates the roulette wheel for our
example. As you can see, each chromosome is
given a slice of a circular roulette wheel.
Selection using roulette wheel
Crossover function
Mutation function
GA cycle
Example 3- 2 variables function