Steps_Up is the procedure that implements our cEA
cea contains the population of our Evolutionary Algorithm. When the algorithm starts, the population is initialized and evaluated.
s is a counter for the number of steps
x is a counter for the number of columns of the population
y is a counter for the number of rows of the population
Compute_Neigh returns a list with the neighbors of the
individual allocated at position (x,y).
Click for more information about neighborhoods.
position(x,y) returns the individual of the population located in the lattice at column x and row y.
selected_inds contains a list of individuals.
Perform_Selection selects one or more individuals from
a list of them -n_list-.
Click for more information about the selection operator.
aux_pop contains the population that is being originated in the current generation.
Apply_Reproduction_Operators combines or modify one or
more individuals to obtain different individual(s).
Click for more information about the reproduction operators.
Evaluate_Population evaluates the solutions represented
by the individuals of the population.
Click for more information about the evaluation functions.
Replace replaces the current population with the population
of individuals created during the last generation.
Click for more information about the replacement policies.
1. proc
Step_Up
(cea)
:
2. for
s
= 1 to MAX_STEPS
do
3. for
x
= 1 to WIDTH
do
4. for
y
= 1 to HEIGH
do
5. n_list
= Compute_Neigh(cea,
position(x,y))
;
6. selected_inds
= Perform_Selection(n_list)
;
7.
aux_pop
= Apply_Reproduction_Operators(selected_inds)
;
8.
end_for ;
9. end_for
;
10.
cea
= Replace(cea,
aux_pop)
;
11.
Evaluate_Population(cea)
;
12. end_for ;
13. end_proc
Step_Up
;
Pseudocode of a simple cEA
Move the pointer over the words or the line numbers for additional information
Next we present a brief explanation with the mean of the main algorithm's lines:
The algorithm runs until one of the two following stopping conditions is satisfied:
In each generation, the algorithm examines all the individuals of the population. This population is hold in a 2-D toroidal grid shape of size WIDTH x HEIGH (more details).
The neighbors of the individual placed on position (x,y) are stored in an empty list.
The parents of the individual of position (x,y) are selected from the list created in line 5 and placed in a new list.
The parents of the individual of position (x,y) are selected from the list created in line 5 and placed in a new list.
The old population is replaced by the new one created in the actual generation.
The individuals of the new population are evaluated.