Mike Tintner wrote:
Brad:>
I presume this is the Waldrop Complexity book to which you referred:
"Complexity: The Emerging Science at the Edge of Order and Chaos"
M. Mitchell Waldrop, 1992, $10.20 (new, paperback) from Amazon (used
copies also available)
http://www.amazon.com/Complexity-Emerging-Science-Order-Chaos/dp/0671872346/ref=pd_bbs_sr_1?ie=UTF8&s=books&qid=1214641304&sr=1-1
Is this the "newer" book you had in mind?
"At Home in the Universe: The Search for the Laws of Self-Organization
and Complexity"
Stuart Kauffman (The Santa Fe Institute), 1995, $18.95 (new,
paperback) from Amazon (used copies
also available)
http://www.amazon.com/At-Home-Universe-Self-Organization-Complexity/dp/0195111303/ref=reg_hu-wl_mrai-recs
Speaking of Kauffman, here's a quote from him, illustrating the points I
was making in the other thread, re how a totally algorithmic approach to
AGI - including an algorithmic trial-and-error approach - won't work
(I disagree with him though - the mind IS a machine, just much more
sophisticated than our current conceptions of machines):
"The second, predominant view among cognitive scientists is that
consciousness arises when enough computational elements are networked
together. In this view, a mind is a machine, and a complex set of
buckets of water pouring water into one another would become conscious.
I just cannot believe this. I cannot however disprove it, but I can
offer arguments against it.
On this view, the mind is algorithmic. With Penrose, in The Emperor's
New Mind, I believe that the mind is not algorithmic, although it can
act algorithmically. If it is not algorithmic, then the mind is not a
machine and consciousness may not arise in a classical - as opposed to
possibly to a quantum - system. Penrose bases his argument on the claim
that in seeking a proof a mathematician does not follow an algorithm
himself. I think he is right, but the example is not felicitous, for the
proof itself is patently an algorithm, and how do we know that the
mathematician did not subconsciously follow that algorithm in finding
the proof.
My arguments start from humbler conditions. Years ago my computer sat on
my front table, plugged into a floor socket. I feared my family would
bump into the cord and pull the computer off the table, breaking it. I
now describe the table: 3 x 5 feet, three wooden boards on top, legs
with certain carvings, chipped paint with the wood surface showing
through with indefinitely many distances between points on the chipped
flecks, two cracks, one crack seven feet from the fireplace, eleven
feet from the kitchen, 238,000 miles from the moon, a broken leaf on the
mid board of the top...You get the idea that there is no finite
description of the table - assuming for example continuous spacetime.
So I invented a solution. I jammed the cord into one of the cracks and
pulled it tight so that my family would not be able to pull the computer
off the table. Now it seems to me that there is no way to turn this
Herculian mental performance into an algorithm. How would one bound the
features of the situation finitely? How would one even list the
features of the table in a denumerably infinite list? One cannot. Thus
it seems to me that no algorithm was performed. As a broader case, we
are all familiar with struggling to formulate a problem. Do you remotely
think that your struggle is an effective "mechanical" or algorithmic
procedure? I do not. I also do not know how to prove that a given
performance is not algorithmic. What would count as such a proof? So I
must leave my conviction with you, unproven, but powerful I think. If
true, then the mind is not a machine.
Stuart A. Kauffman , BEYOND REDUCTIONISM, Reinventing The Sacred, Edge,
11.13.06, http://www.edge.org/3rd_culture/kauffman06/kauffman06_index.html
What Kauffman is talking about here is the "Frame Problem". Anyone who
has gone through a standard AI/Cognitive Science training should
recognize that.
But now here is the trouble with this argument. What does he mean by
saying that the mind is not 'algorithmic'? He uses the keyphrase
'effective procedure' when trying to describe this, but that is a loaded
techical term....
What he means by 'algorithm' in this context is what some of us would
call the rigid manipulation of simple, hard-edged symbols, using metods
that have explicit semantics.
BUT if you go outside that interpretation of 'algorithm' and include
mechanisms that work by a process of dynamic, stochastic relaxation, it
is easy in principle to see how this issue (the Frame Problem) could be
solved. Or rather, it becomes difficult to see that a problem actually
exists at all.
The trouble is, that many of us would say that dynamic relaxation is
just as algorithmic as anything else. It just does not involve symbols
and mechanisms closed-form, explicit semantics. There is no big
mystery here, no destruction of the Computational Paradigm. It is just
a different way of looking at what 'algorithm' means, that's all.
Richard Loosemore.
-------------------------------------------
agi
Archives: http://www.listbox.com/member/archive/303/=now
RSS Feed: http://www.listbox.com/member/archive/rss/303/
Modify Your Subscription:
http://www.listbox.com/member/?member_id=8660244&id_secret=106510220-47b225
Powered by Listbox: http://www.listbox.com
