6 ms·

The point of genetic algorithms is NOT to replicate evolution. Like you mention, evolution has no end goal whereas engineers usually have very explicit end goal

The point of genetic algorithms is NOT to replicate evolution. Like you mention, evolution has no end goal whereas engineers usually have very explicit end goals. (Other problems include time constraints and computing power!) The point of genetic algorithms is to exploit the principle of natural selection under very specific constraints and explicitly constructed utility functions that are likely to lead to a better solution to a given problem. Of course we're cheating a little, but we have to. "Organic" genetic algorithms that consistently achieve engineering design principles (like robustness, modularity, compartmentation, etc.) without constraints are the holy grail of the field.

kirse 18y ago
+1 Thanks for the explanation. I can fully understand both the legitimacy and the desire to model the specific process of genetic propagation, but was somewhat confused how exactly that logically validates it as an accurate representation of its function in evolution as a whole. It's pretty clear one model has a designer and the other does not, and you certainly agree that the existence of a designer is more than just a small issue when dealing with the validity of evolutionary "models". =)
- demallien 18y ago
  If you mean that there is a 'designer' of the system, then sure, there is. But if you mean that there is a 'designer' of the little model cars, then no, you are absolutely wrong (except in the sense that the GA itself, a mindless mathematical process, is a designer). In other words, this is a pretty faithful model of evolution if we postulate that once the first self-replicating organism came into being (once the programmer had written the model/fitness function), no further interference is necessary - a purely mathematical process will lead to artifacts that appear 'designed'.
wheels 18y ago
To be clear, the field of genetic algorithms is split into two groups: one primarily concerned with function optimization for engineering application, the other at providing a laboratory for understanding evolutionary processes.
demallien 18y ago
I beg to differ - evolution, in the 'life on Earth' sense has a very very specific goal - to reproduce genes as successfully as possible. To date, we know no life that does not seek this primal goal.
- kirse 18y ago
  You clearly misunderstand your own theory, because this specific paper written by an evolutionist (on TalkOrigins, the shrine of evidence for Evolution) reinforces the view on why biological evolution has no goal: http://www.talkorigins.org/faqs/evolphil/teleology.html http://www.talkorigins.org/faqs/evolphil/teleology.html When Dawkins talks about genes maximising their representation in the gene pool, this is a metaphor not an explanation. Genes just replicate. It happens that those that out-replicate others end up out-surviving them. There is no 'goal' to genetic behaviour. Evolutionists are always screaming about the evidence, and it's rarely been debated that biological evolution is completely devoid of a designer, purpose, intelligence, or end goal anywhere in the process. That's one of the basic truths you get out of the naturalistic assumptions that base biological Evolution.
  - demallien 18y ago
    1) It's not my theory 2) Whilst it is true that individual genes do not have goals, it is not true that natural evolution, as a search algorithm, does not. When we talk about goals for GAs (which is of course the context of this very discussion we are having), we are specifically talking about high values of the fitness function of the GA. In the case of natural evolution, the 'goal' is to reproduce a gene as much as possible. Note, this is a goal of the GA, not of the gene, and the usage of the word here does not indicate an actually entity with a motivation, but simply reuses terminology that is quite common in discussion of search algorithms (evolution being but one example of these).

kirse 18y ago

+1 Thanks for the explanation. I can fully understand both the legitimacy and the desire to model the specific process of genetic propagation, but was somewhat confused how exactly that logically validates it as an accurate representation of its function in evolution as a whole. It's pretty clear one model has a designer and the other does not, and you certainly agree that the existence of a designer is more than just a small issue when dealing with the validity of evolutionary "models". =)

demallien 18y ago
If you mean that there is a 'designer' of the system, then sure, there is. But if you mean that there is a 'designer' of the little model cars, then no, you are absolutely wrong (except in the sense that the GA itself, a mindless mathematical process, is a designer). In other words, this is a pretty faithful model of evolution if we postulate that once the first self-replicating organism came into being (once the programmer had written the model/fitness function), no further interference is necessary - a purely mathematical process will lead to artifacts that appear 'designed'.

wheels 18y ago

To be clear, the field of genetic algorithms is split into two groups: one primarily concerned with function optimization for engineering application, the other at providing a laboratory for understanding evolutionary processes.

demallien 18y ago

I beg to differ - evolution, in the 'life on Earth' sense has a very very specific goal - to reproduce genes as successfully as possible. To date, we know no life that does not seek this primal goal.

kirse 18y ago
You clearly misunderstand your own theory, because this specific paper written by an evolutionist (on TalkOrigins, the shrine of evidence for Evolution) reinforces the view on why biological evolution has no goal: http://www.talkorigins.org/faqs/evolphil/teleology.html http://www.talkorigins.org/faqs/evolphil/teleology.html When Dawkins talks about genes maximising their representation in the gene pool, this is a metaphor not an explanation. Genes just replicate. It happens that those that out-replicate others end up out-surviving them. There is no 'goal' to genetic behaviour. Evolutionists are always screaming about the evidence, and it's rarely been debated that biological evolution is completely devoid of a designer, purpose, intelligence, or end goal anywhere in the process. That's one of the basic truths you get out of the naturalistic assumptions that base biological Evolution.
- demallien 18y ago
  1) It's not my theory 2) Whilst it is true that individual genes do not have goals, it is not true that natural evolution, as a search algorithm, does not. When we talk about goals for GAs (which is of course the context of this very discussion we are having), we are specifically talking about high values of the fitness function of the GA. In the case of natural evolution, the 'goal' is to reproduce a gene as much as possible. Note, this is a goal of the GA, not of the gene, and the usage of the word here does not indicate an actually entity with a motivation, but simply reuses terminology that is quite common in discussion of search algorithms (evolution being but one example of these).