Not only do I use them.... I make them.
And my customers paid me a lot of money to make them as well.
Well I am very pleased that you are monetarily successful in your business.
It is a blind process.
In the sense that:
- changes aren't planned
- changes aren't predetermined
- selection only looks at fitness of the CURRENT generation (ie: it doesn't go like "hmmm, I'm gonna keep this one cause it might be nice to have it 20 generations down the line").
In that sense, it is completely blind.[/Quote]
I haven't implied that any of the above are planned. You misunderstand my objections to the GA. They incorporate intelligence in the program by the knowledge we put into them, that is one issue and although they are random this is based on optimization.
Secondly and most importantly, it is not a true representation of biological evolution. That is where you are not understanding the issue.
The actual point is that the appearance of deliberate design is being produced by a mindless process without any intelligent intervention.
No. It is designed to evolve solutions for design problems.
It is only designed insofar as to create an environment where this process can take place.
Just like a freezer is designed to create an environment where the process of freezing can take place.
Freezing is not the result of design.
Neither is the process of evolution.
Freezing / evolution is rather a thing that inevitably happens if the environment finds itself in a specific state with certain properties.
In case of evolution, one needs systems that compete and reproduce with variation and heredity, and are subject to some kind of fitness test in context of the "competition".
What is designed in a GA is that environment. Just like a freezer.
...to create the software.
The designs that the software comes up with is the result of a blind process executed by the algoritm the software uses.
Now I want to present another person who is not a theist, is a biochemist and who understands evolution and who has no motivation in this discussion:
WEDNESDAY, JUNE 17, 2015
A recent article in
Nature reminded me of the importance of definitions. The article discusses evolution and evolutionary algorithms in a special issue on machine learning (Eiben and Smith, 2015). I think we all know that "evolutionary" algorithms are based on natural selection and we all know that there's more to evolution than just adaptation. It's too late to change the name of these procedures in computer science but at the very least I expect computer scientists to be aware of the difference between their procedures and real evolution.
In this paper, there's a section on "how evolutionary computation compares with natural evolution." The authors consistently use "evolution" as a synonym for "selection" or "adaptation" and they seem to be unaware of any other mechanism of evolution.
In one sense, it's okay to conflate "evolution" and "adaptation" in computer science but if that error reflects and perpetuates a fundamental misunderstanding of the nature of real biological evolution then perhaps it's time to rename these algorithms "adpatation algorithms."
http://sandwalk.blogspot.com/2015/06/evolutioanary-algorithms-are-really.html
Learn about evolution.