Hi Edward,

 

Haven't figured out how to get rid of the HTML line at the side in Outlook
so I'll reply at the top here.

 

Our heads are doing pure and absolute they are busily cranking away at it.
Our heads are also an "instance" and a subset of a pure and absolute. 

 

Gosh I suppose I have to back that up huh? Well it depends on the pure and
absolute reference point. An easy answer is to take the data or digital
physics perspective and say all matter is fundamentally made of data and/or
has a data representation at its minute granular level. Can you run with
that for a while? J As John quickly changes the subject since this is going
to require lot's of explanation.

 

CAs? Seems like you are well versed on them. I had mentioned them as they
are always coming into play as very useful tools and simulators of .well.
potentially everything, but using them for higher level constructs like
FSMs, pattern generators, logic synthesis, chaos processing,.. I have been
evaluating them as alternatives for current programming constructs. As to
whether they are efficient don't know and actually have been trying to avoid
them. Wolfram seems to like them. they have "core" properties.

 

John

 

 

From: Edward W. Porter [mailto:[EMAIL PROTECTED] 



John, 

 

Thanks for the pointers.  

 

I took a quick look at both your links.  Neither of them seem to have the
mindset about AI that I have -- which, if one was going to divide all
approaches to AI into ten bins, would probably put my thinking and in the
same bin as Novamente.  And I am trying to get some fairly heavy AGI reading
done that is more along the lines of thinking in that bin.

 

With regard to algebraic structures -- I just skimmed the wikipedia page on
that subject. I can't really criticize it because I haven't taken the time
to understand it.  But when I see a lot of lingo from rigorous math and
logic, I start thinking it doesn't sound like my concept of AGI, which is
focused on experiential intelligence, intelligence based on memory, patterns
derived from memory, probability, and flexible context adjustable patterns
and metrics of similarly.  Rather such lingo suggests the realm of the pure
and the absolute - of mental fascism.  As I have said for years rigorous
logic is to human thought what dressage is the motion of horses.  Except in
its simplest forms, it is unnatural.

 

My head doesn't do pure and absolute.  And, in fact, probably neither does
yours.

 

So that is just an ill informed gut reaction from a few minutes of reading
the wikipedia algebraic structure link you sent me.  If I am way off base,
which is highly possible, please inform me.

 

With regard to cellular automata, it depends how you define them.  I very
much believe in a distributed architecture with relatively simple local
processing and messaging.  That is what neurons do.  And it scales well on
massively parallel architecture.  So if that falls within the bailiwick of
cellular automata, cool.

 

But if by cellular automata, you mean that a node can only compute through a
limited number of guy's next to it in a relatively low dimensional space,
fagedaboutit.  

 

(I am not, however, necessarily opposed to grid or toroidal networks, I am
just saying that except at lower motor or perceptual levels, the
architecture should be designed to efficiently handle one hell of a lot of
non-local messaging.) 

 

I am interested in computing world knowledge and world knowledge is best
thought of as residing in a sparsely filled very high dimensional space, in
which the dimensions are constantly warping (a la Hofstadter's slipnet) and
in which determining what is a nearest neighbor node is often highly
confusing and variable.

 

So again, if cellular automata would cover what neurons do, they have a lot
of promise, although I would not limit myself to just processing on such a
low level.  But if cellular automata are limited to a notion of you do all
your communication through computations with in a limited set of neighbors
defined in anything less that a space with well over a million dimensions
(see my , fagedaboutit.

 

Please tell me if I am wrong, and which of these two interpretations
cellular automata is the standard one.

-----
This list is sponsored by AGIRI: http://www.agiri.org/email
To unsubscribe or change your options, please go to:
http://v2.listbox.com/member/?member_id=8660244&id_secret=55828852-bde5af

Reply via email to