References about Galapagos?

Hi, david and everyone...

These days I'm studying basics of Genetic Algorithms and Galapagos....

And I'm trying to apply Galapagos to optimization of structures with some success.

However, the technical terms in galapagos component is not so easy to undertand.

There are so many varied genetic algorithms, so they are confusing...

Can I get a list of reference documents or dissertaions about the Genetic Algorithm applied to Galapagos? Or can I get some detailed document on Galapagos?

Replies to This Discussion

Permalink Reply by Cristiano_010 on September 23, 2010 at 1:38am

Hi, I am studying the behavior of Galapagos and relating it with genetic algorithms.
information galapagos final delivery after finding the optimal solution is as follows

genome [37], fitness = 21.00, Genes [40% * 65% * 67%]
record: Genome WAS mutated To Avoid collision?

I wonder that means each of these points, I have an idea that genes would come to be given parameters.
fitness evaluation is a function of how it fits the possible solution to the problem.
genome would be the solution obtained after the crossing.
I hope to help me get more information about it or tell me if I'm wrong.
thanks

Cristian Namoncura
cnamoncur@gmail.com

Permalink Reply by omar_etman2000@hotmail.com on April 29, 2013 at 5:42pm

hello David

i am working with Galapagos on this project i just needed to know what does the graphs repesnt

the point graph in the bottom right and the line graph next to it and the generation graph

be grateful if you can help me with this

Omar etman

Permalink Reply by VALENTINA on February 16, 2018 at 2:25am

Hi David, thak you for you explanation.

I have another doubt: how it is possible to set the type of coupling (isotropic or exclusive) and how a type of coalescence (crossover, interpolated or preference blending) in Galapagos? How could I know which option is the solver employing?

Thank you very much!

Valentina

Permalink Reply by Naama Glauber on November 14, 2018 at 10:08am

My Question:

Hi David,

I really enjoyed your detailed explanation about Galapagos!As a computer scientist- for me it is matter how things have been done behind the scene :) Grassopper is really cool and helpfull for me, and I always wanted to know and explore it..

I used Galapagos for solving some task, and now I'm writng an article about what I'm doing..
I have several questions regarding the algorithm's steps you mentioned (I hope you can answer):

In your explanation you described several options for some parts of the algorithm (how to make coupling, mutating etc..). Can you please explain more in detailed (parameter, or atleast the methods only) you used for Galapagos?

To be persicely:

what is the population size on the beginning?

5.a) Did you use isotopically, exclusively or biased??
- If exclusively, what percentage?
- If biased what is the 'vector of weights'? or however you implemented that..

5.b) For the implementation- do you have some Gaussian with a pick on the 'Inbreeding factor' (which is some number in [0,1], while 0 presents 'incestuous', and 1 presents 'zoophilic' or the opposite)?

5.c) Did you interpolated the values by averaging (i.e. equal weights) or using preference weight according to the fitness?

5.d) I see What you said about number of sliders, want to be sure I understand: the mutation here is just to pick some percentage of the genes (what percentage you used?) and change the child's value to be a random number in the range of the slider?

Can I change the percentage of individuals from G[n] are allowed (you said the default is 10%)?
What is the default for this? Is it the first one reached?

Can I specify the max number of iteration?
Can I specify the number of generations?
Can I specify a fitness value to stop in?

Maybe I missed some parameters, but I saw Galapagos as a "black box". But maybe I missed I can adjust it (in the later case, would like to know what are the defaults values).

I guess it is not an open-source code (right?) and maybe you don't want to share it for public. I would be glad if possible to know a bit more at least about the methods, so I describe it when writing my article, please J
You can also answer me here: naama.glauber@gmail.com

Permalink Reply by Mårten Nettelbladt on July 16, 2010 at 5:56am