Jim,

 

I read this thread and grow more and more alarmed. Everything that happened
in Aeronautical Engineering was tied to a principle. A principle is a
mathematical statement that descibes a law of nature. A principle can be
used to engineer other things, such as AGI or airplanes, because it is
sufficiently mathematical for that purpose, but can not itself be engineered
because it is natural. That's what makes it a principle. 

 

Aeronautical Engineering started when the forces of lift, drag and
propulsion were explained in terms of Newton's equations. That is the
principle. Knowing this principle, useful replication became possible
because the space of possibilities had been sufficiently narrowed down, and
everything else being said in this thread followed. Without the principle,
we would be allowed to believe that the feathers of birds are "creative" or
"intelligent" in some way because they can "reason" and they "know" how to
make the bird fly. 

 

In AGI, there is no principle. For some time now I have been proposing
emergent inference (EI) as a principle for AGI. EI is a mathematical
statement, it is mathematical enough to be used for engineering AGI, and it
can not be engineered itself because is is a natural law. There is evidence
that EI, and what we today call intelligence, have something to do with each
other. Furthermore, the mechanism of EI is such that it would not be
discovered even if a full synapse-to-synapse replication of the entire brain
were made. Even in that case, EI would remain hidden. Brain replicators and
reverse-engineers are not after EI because they don't know EI exists. 

 

I feel EI must be seriously considered. 

 

On a different note, the Superconducting Supercollider (SSC) was not
intended to replicate anything, except of course that testing it would have
required to replicate many experiments that had been already performed, to
make sure it works. The SSC was intended to observe the Higgs boson. I
should know. I worked on the SSC for years until it was cancelled by
Congress. And yes, physicists miss the observer, but not totally, they just
don't know what to do about the observer. 

 

Sergio

 

 

 

From: Jim Bromer [mailto:[email protected]] 
Sent: Tuesday, June 26, 2012 10:53 AM
To: AGI
Subject: Re: [agi] Happy 100th Birthday Alan Turing - No, computers will
never think, but machines will!

 

On Tue, Jun 26, 2012 at 2:15 AM, Colin Geoffrey Hales
<[email protected]> wrote:

Physicists get the need for replication, but totally miss the need for the
observer in science.

Neuroscientists are examine the physics of the observer, and total miss the
role of replication in science.

Sheesh this is hard.

 

That is an interesting point but you are dealing with two areas on the
frontiers of science.  It is a little too conservative.  You want to be on
the edge of new sciences but wish that they were framed with the dark
stained oaks of the finest institutions.  You sound a little like a
caricature of a somewhat stuffy academic.

 

In replying to you I did see that simple simulations that led to basic
insights about how an airplane should be designed were a key part of the
Wright's method of development so I am going to start doing some simple
experiments in AGI.  However, the argument that physicists do not get the
need for observer and neuroscientists do not get the need for replication is
not insightful.  I don't know why the elimination of the necessity of
working from a foundation of observed physical effects lost traction with
some physicists, however, the popularization of the concept signifies the
trivialization of the concept.  The concept has been trivialized.

 

Jim Bromer 

 


 

AGI |  <https://www.listbox.com/member/archive/303/=now> Archives
<https://www.listbox.com/member/archive/rss/303/10561250-164650b2> | Modify
Your Subscription

 <https://www.listbox.com/member/archive/rss/303/10561250-164650b2> 

 <https://www.listbox.com/member/archive/rss/303/10561250-164650b2>  




-------------------------------------------
AGI
Archives: https://www.listbox.com/member/archive/303/=now
RSS Feed: https://www.listbox.com/member/archive/rss/303/21088071-c97d2393
Modify Your Subscription: 
https://www.listbox.com/member/?member_id=21088071&id_secret=21088071-2484a968
Powered by Listbox: http://www.listbox.com

Reply via email to