matlab program for comparison and loop

Chromosome[1] = [12;05;23;08] Chromosome[2] = [02;21;18;03] Chromosome[3] = [10;04;13;14] Chromosome[4] = [20;01;10;06] Chromosome[5] = [01;04;13;19] Chromosome[6] = [20;05;17;01]

R[1] = 0.201 R[2] = 0.284 R[3] = 0.099 R[4] = 0.822 R[5] = 0.398 R[6] = 0.501

C[1] = 0.1254 C[2] = 0.2710 C[3] = 0.4118 C[4] = 0.6639 C[5] = 0.7882 C[6] = 1.0

i want to write a simple matlab program in which i have to compare the r(k) with value of c(k) if r(k) lies within the range of c(k) then it will return chromosome of that particular k. actually it is part of genetic algorithm i have write 30% program but i am confused at roulette stage how to write it.

the response of the above program must return the following output.

Chromosome[1] = [02;21;18;03] Chromosome[2] = [10;04;13;14] Chromosome[3] = [12;05;23;08] Chromosome[4] = [20;05;17;01] Chromosome[5] = [10;04;13;14] Chromosome[6] = [20;01;10;06]

manual calculation are as follow.

as r1=0.201 doesnot lie in the range of c1=0.125 (c1<r1), so we move forward and compare it with next, means with c2. as r1=0.201<c2 = 0.2710 it means it must replace chromosome of 1 with 2.

following is the table of replacement.

NewChromosome[1] = Chromosome[2] NewChromosome[2] = Chromosome[3] NewChromosome[3] = Chromosome[1] NewChromosome[4] = Chromosome[6] NewChromosome[5] = Chromosome[3] NewChromosome[6] = Chromosome[4]

[Mudasir Ahmed]

kindly help me in this regard :)

[Niraj Reginald]

Dear Ahmed, Could you share you code to [e-mail] Thanks!

[Niraj Reginald]

Dear Ahmed, Could you share you code to [e-mail] Thanks!

Answer 1

[Geoff Hayes]

Mudasir - from your roulette selection scheme to select parents, where the vector R seems to be the randomly assigned values (between 0 and 1) for each parent, and C is the cumulative probability distribution vector, then you can implement the following to get the parent selection

 R(1) = 0.201;
 R(2) = 0.284; 
 R(3) = 0.099;
 R(4) = 0.822; 
 R(5) = 0.398; 
 R(6) = 0.501;

 C(1) = 0.1254; 
 C(2) = 0.2710;
 C(3) = 0.4118;
 C(4) = 0.6639;
 C(5) = 0.7882; 
 C(6) = 1.0;

 for k=1:length®

   % use find to find the first (1) element of
   % C that is greater than R(k)
   idx = find(C>R(k),1);

   if isempty(idx)
       idx = length(C);
   end

   % select chromosome idx as new parent

 end

The above determines the idx according to R and C; you just need to map the population chromosomes to the new parent set.

[Mudasir Ahmed]

thanks sir :).

http://arxiv.org/ftp/arxiv/papers/1308/1308.4675.pdf

sir i am writing a program for genetic algorithm by considering a example solved in above link. i successfully write at a stage of crossover, but now confused at mutation stage. bcz in text author write contradictory statements like: i am pasted below.

"""""""""""""""""""""""" Number of chromosomes that have mutations in a population is determined by the mutation rate parameter. Mutation process is done by replacing the gen at random position with a new value. The process is as follows. First we must calculate the total length of gen in the population. In this case the total length of gen is total_gen = number_of_gen_in_Chromosome * number of population = 4 * 6 = 24 Mutation process is done by generating a random integer between 1 and total_gen (1 to 24). If generated random number is smaller than mutation_rate(ρm) variable then marked the position of gen in chromosomes . Suppose we define ρm 10%, it is expected that 10% (0.1) of total_gen in the population that will be mutated: number of mutations = 0.1 * 24 = 2.4 ≈ 2 Suppose generation of random number yield 12 and 18 then the chromosome which have mutation are Chromosome number 3 gen number 4 and Chromosome 5 gen number 2. The value of mutated gens at mutation point is replaced by random number between 0-30. Suppose generated random number are 2 and 5 then Chromosome composition after mutation are:

Chromosome[1] = [02;05;17;01] Chromosome[2] = [10;04;13;14] Chromosome[3] = [12;05;23;02] Chromosome[4] = [20;04;13;14] Chromosome[5] = [10;05;18;03] Chromosome[6] = [20;01;10;06]

""""""""""""" as in above text author says that, he takes mutation rate 10% (0.1)and he also mention that if random number generated between (1 to 24) are less then mutation rate then we have to mark the position. and replace with another random number generated between (1 to 30). as numbers generated were 12 and 18 which are greater then 10% or 0.1 then how he mark position 12 and 18 and replaced with 2 and 5. i am totally confused sir at this stage. kindly help me

[Geoff Hayes]

Mudasir - looking at his paper (in the link you provided), the statement which you highlighted in bold does not make all that much sense. I think that you can ignore it. He may be alluding to how other mutation processes work where you mutate the gene only if the random number generated for that gene is less than the mutation rate. But when he states that Mutation process is done by generating a random integer between 1 and total_gen (1 to 24). If generated random number is smaller than mutation_rate(ρm) variable then marked the position of gen in chromosomes, it is unclear how an randomly generated integer between 1 and 24 can ever be less than the mutation rate, which will always be less than one!

I think if you follow his next steps and just realize that if the mutation rate is 10%, then 10% of the genes in the whole population should be mutated. So this is just

 numGenes = 4;    % number of genes per chromosome (number of variables)
 numChrom = 6;    % number of chromosomes (population size)
 mutRate  = 0.1;  % mutation rate
 totalNumGenes = numGenes*numChrom;  % total number of genes 
                                     % across all chromosomes
 numGenesToMut = floor(0.1* totalNumGenes); % number of genes to mutate

So there are two genes to mutate, from numGenesToMut. And so we must randomly generate two integers from 1 to 24. In his example, the random numbers are 12 and 18. If we consider 12, since there are 4 genes per chromosome, then the mutation occurs in the 4th gene of the 3rd chromosome as he has shown above (since 12=4+4+4). With 18, again since there are 4 genes per chromosome (imagine all 6 chromosomes lined up, side by side), then the mutation occurs at the second gene of the fifth chromosome (since 18=4+4+4+4+2).

[Mudasir Ahmed]

sir i successfully make a program for Genetic algorithm (without using elite feature) and achieve goal which was desired. i have tried for various number of chromosomes like.5,6,10 and 20. but didn't notice any significant change. sir as i logically believe, there must me some impact on output response by increasing the number of chromosome. and sir my second question is i want to know, at which stage elite strategy is used and how. i am following the below link: http://www.mathworks.com/help/gads/how-the-genetic-algorithm-works.html in which a line is written "Some of the individuals in the current population that have lower fitness are chosen as elite. These elite individuals are passed to the next population." i am confused at stage , means if a individual have low fitness then how it is processed to a next level?.

kindly share information if you have any, thanks in advance sir