News for hackers. Stuff that matters.
Re:Genetic Algorithm building a little car by kirse on Monday December 08, 2008 @04:03PM (#390372)
How exactly is it evolutionary computing when there is a designer in the equation (programmer) who puts constraints on the algorithm by the simple act of coding it and running it on systems that he designed?

This seems more like "computing to model the process of genetic adaptation". You can't label it evolutionary when there's a designer in the equation. The very definition of the evolutionary process does not allow for constraints on the process itself nor an end goal/purpose, something inherent to any system built by a programmer.

Re:Genetic Algorithm building a little car by malkia on Monday December 08, 2008 @04:11PM (#390386)
You can think of the programmer as the physics laws in the universe.

This law, being a law, just dictates rules of what is possible, but the outcome is what happen in the mean-time.

Some people do think that God is the Physics Law. Some do think that God is the Time, where this happened.

Re:Genetic Algorithm building a little car by kirse on Monday December 08, 2008 @04:26PM (#390404)
If the programmer specifies the boundaries (laws) of his own universe, then the evolutionary processes that happen within his own universe are not representative of the actual universe which he claims to be modeling. Why?

1) Current scientific claims express that the laws and boundaries of the universe came into existence without a designer. Having a designer anywhere in the equation is unacceptable, even one who simply created the universe, because that invalidates (or at least casts heavy doubt) on the naturalistic claims of any designer-free subprocesses that happen within the universe (specifically, Evolution).

2) Evolution itself is devoid of a designer and end goal.

Logically, it is impossible to model a process that is inherently devoid of any design or purpose.

It would only be scientific to question the accuracy of any model that can't even replicate the basic core constraints (eg. lack of a designer) of the system which it is trying to represent.

Re:Genetic Algorithm building a little car by nsrivast on Monday December 08, 2008 @04:35PM (#390412)
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.

Re:Genetic Algorithm building a little car by wheels on Tuesday December 09, 2008 @04:06AM (#391125)
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.
Re:Genetic Algorithm building a little car by kirse on Monday December 08, 2008 @04:39PM (#390418)
+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". =)

Re:Genetic Algorithm building a little car by demallien on Tuesday December 09, 2008 @04:57AM (#391145)
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'.

Re:Genetic Algorithm building a little car by demallien on Tuesday December 09, 2008 @04:52AM (#391143)
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.
Re:Genetic Algorithm building a little car by kirse on Tuesday December 09, 2008 @12:50PM (#391496)
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

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.

Re:Genetic Algorithm building a little car by demallien on Wednesday December 10, 2008 @05:12AM (#392725)
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).
Re:Genetic Algorithm building a little car by DavidSJ on Monday December 08, 2008 @09:00PM (#390786)
Your definition of evolution is incorrect. Evolution is a process which occurs in any system that has:

* replication

* selection pressure

* variation

No claims about that system's progeny are made by evolution.

Re:Genetic Algorithm building a little car by RK on Tuesday December 09, 2008 @01:37PM (#391555)
Evolution simply means change over time.

Your description is more akin to natural selection as a mechanism for evolution.

Re:Genetic Algorithm building a little car by DavidSJ on Tuesday December 09, 2008 @01:59PM (#391587)
Words have multiple definitions. Yes, "evolution" simply meant change over time, prior to Darwin's publication of the Theory of Evolution. In the context of this conversation, "evolution" refers to the process Darwin was describing, but in reference to digital individuals rather than biological organisms.

Certainly, the word "evolution" in none of its permutations describes how the process started.

Re:Genetic Algorithm building a little car by anthonyb on Monday December 08, 2008 @07:08PM (#390636)
Do you think evolution really cares whether constraints are artificial or not?
Re:Genetic Algorithm building a little car by kirse on Monday December 08, 2008 @08:35PM (#390751)
I knew someone was going to post this and should have addressed it..

Programming an evolutionary model is not setting artificial environmental constraints and then letting an unbounded process operate at will (as your comment implies). You would agree that the very act of coding a process puts constraints upon it. Programming any evolutionary model puts artificial constraints on both the environment and the evolutionary process itself.

Regardless, as "nsrivast" cleared up above, they are not striving to model Evolution (nor should they be claiming a process that has no constraints, intelligence or purpose is modeled accurately via constraints, a designer, and an end goal). As nsrivast noted, some of the most common constraints are time and computing power.

Re:Genetic Algorithm building a little car by anthonyb on Monday December 08, 2008 @10:55PM (#390921)
I think the quote that you're looking for is: "All models are wrong; some are useful".

That said, selection pressure is selection pressure, regardless of whether the means are artificial (being culled from a list in RAM) or natural (being eaten by a lion). I don't see that evolution in a computer process, even with artificial constraints, is that much different to evolution in the "real world".

I'm also not 100% sure exactly what your definition of an artificial constraint is, but something like Tierra might be more what you're looking for.