Hi @milad.showkatbakhsh and Dear community
I want to optimize energy and daylight according to different facade geometrical variables.
my no. of genes are 5 and no. of values are 1024 hence I want to know, what is the best generation size and count for my problem. Also, my question is how can we explain the crossover and distribution indices that are appropriate for the proposed problem too.
Please take a look at this post where @mmakki_10 responded to a similar question.
And regarding the indices, based on our experience, this depends on the problem at hand, So what I suggest is to run the simulation with default indices. And then analyse the results, and if the analysis indicated that exploration or exploitation need to be adjusted (i.e. if you spotted early convergence), then you can adjust the indices. The indices and their impact on the simulation are explained in our primer as well.
Thanks, Milad for what you said, and fortunately I have read the primer. that’s really useful but here is another question. Is the first generation of wallacei (NSGA-II) has 2x pop size of other generations like octopus?
No, the first pop size (generation size in Wallacei) is not 2X. Another piece of information that may help: In every generation, all the parents and all the offspring compete with one another.
Many thanks for your help and that pops up another question, the competing part is one of the main differences between NSGA-II and PEA-2 or HypE?
Obviously, this would help me out a lot to know more specific differences in the comparison of these algorithms behind the scene.
There is alot of literature online about the differences between NSGA2 and SPEA2 (as well as other MOEAs)… id recommend you first read the papers published by the authors of these algorithms, followed by the various comparative analyses published by third parties… here are a few links to get you started:
Papers by original authors:
A Fast Elitist Non-Dominated Sorting Genetic Algorithm for Multi-Objective Optimization: NSGA-II:
A Fast and Elitist Multiobjective Genetic Algorithm: NSGA-II
SPEA2: Improving the Strength Pareto Evolutionary Algorithm:
Survey of Multi Objective Evolutionary Algorithms:
A Summary and Comparison of MOEA Algorithms
A Comparison of MOEA/D, NSGA II and SPEA2 Algorithms:
theres much more… a quick search on google scholar will reveal more.
Many thanks @mmakki_10
I’ve read some of the articles but I’m pretty sure the ones that you recommended are much more helpful.