From: Walter Hicks (wallyshoes@mindspring.com)
Date: Sat Nov 01 2003 - 16:49:42 EST
"D. F. Siemens, Jr." wrote:
> On Fri, 31 Oct 2003 20:51:21 -0500 Walter Hicks
> <wallyshoes@mindspring.com> writes:
> >
> >
> > bivalve wrote:
> > <snip>
> > > Engineering is looking for optimization, and thus resorts to a
> > strategy similar to that promoted by Intelligent Design advocates.
> > Evolution only requires good enough; rigorous selective pressure
> > would be necessary to approach optimization, and even then it is
> > contingent on historical possibilities and other constraints.
> > E..g., humans might find a third grasping appendage handy, but tails
> > were lost in the ape/human lineage (and were not prehensile in that
> > group anyway). This historical constraint limits human evolution,
> > but an engineer could simply graft something on. Raising the
> > mutation rate in the hopes of growing another appendage would
> > probably be more detrimental due to harmful mutations than effective
> > towards the goal, and so other constraints come into play.
> >
> > Yes, but as I said, the basic point is to imitate nature because
> > nature has been so effective in getting :the "best": solutions to a
> > given parameter (not just a useful one ----- and this is indeed the
> > case. That is why DARPA clings to the approach. If the evolutionary
> > model were correct, then GA should indeed get the best answer for
> > parameter X, just as nature does. When radar was invented, we found
> > out that bats were way ahead of us. I believe that they still are.
> > We cannot make devices that smell as well as a dog can, etc. If the
> > biological model were correct, then GA would produce far more than
> > the rough answer that it currently produces.
> >
> > <snip>
> > Walt
> >
> I'm trying to figure where you got the idea that nature via evolution
> either produces optimum solutions, or the theory is erroneous.
I don't understand the "either .. or" nature of the above sentence. What
nature does is one thing and what the theory does is the question. I did not
say anything like the above.
> Survival
> does not direct change, but only sorts out what is available within the
> restrictions of the specific environment--with lots of room for chance
> events that have nothing to do with fitness. A different environment
> would probably show a different combination to be most fit. Further, the
> environment is variable to a greater or lesser extent. There is no way to
> maximize the variety of genotypes for testing in natural circumstances,
> and the test at best is of phenotypes.
True. That is what nature does. My issue is with the theory of how that
happens.
> All this is very different from
> the human use of GA, which generates a maximum number of variants in a
> protected environment and tests for desired results. As I understand the
> situation, if we totally understood all the factors involved, we could
> design and construct an enzyme, say, that would do the required job most
> efficiently. But this would be worse than solving a random walk problem,
> for the specific goals within the problem are not set. Lacking
> understanding, we throw everything we can at the problem and sort out
> results--a shotgun approach.
>
> To bring down a flying duck with minimum expenditure of energy, a .22
> short through the head is probably the best. The .177 air rifle pellet
> does not have the power to fly far enough and hit hard enough. However,
> duck hunting with a .22 is futile, for no one can aim accurately enough.
> Even assuming that the trajectory could be calculated exactly and the
> muzzle pointed to microseconds, variations in the charge and its burning,
> and in the bullet, plus whatever other variables might enter in, would
> make hunting ducks with a .22 stupid. A shotgun load covers a large
> enough area for the hunter to have a reasonable chance of downing a duck,
> with a much larger expenditure of energy and no guarantees. None of this
> demonstrates that rifles are non-functional.
> Dave
I return to what I have said and elaborate further.
I have friends at DARPA and I have attended some of DARPA's presentations
where a favorite theme is --- given a problem --- look at what nature has
done. For the most part, nature has arrived at a solution that is much much
better than what our engineering tools ever generate. Accordingly, DARPA
often looks to nature for solutions. I see GA used in some DARPA programs for
exactly that reason.
So where do these GA tools some from.? They come for current biological
theory (like yours, Dave). In fact, I notice that some GA books show their
use in the biological field. Wally Hicks did not make all this up. It is the
work of numerous other people.
Now, if the tools fall short of accomplishing what nature does, then it is
fair to say (I think) that the tools do not yet replicate nature -- because
the underlying theory is not yet adequate to do so. What on earth is wrong
with that?
I really do not understand why one can accept the fact that evolutionary
theory is a "work in process" and then get in a snit when it is pointed out
that it does yet work. I only say all this because there are many who like to
present evolutionary theory as if it were a "done deal" and that leaves me
rather cold ---- brrrrrrrr.
Walt
-- =================================== Walt Hicks <wallyshoes@mindspring.com>In any consistent theory, there must exist true but not provable statements. (Godel's Theorem)
You can only find the truth with logic If you have already found the truth without it. (G.K. Chesterton) ===================================
This archive was generated by hypermail 2.1.4 : Sat Nov 01 2003 - 16:52:52 EST