Re: Paper by ~MM on distributed self-awareness



SH: I used google to investigate that man other people had
written about self-awareness and emergence. Her usage
seems to be fairly standard. Maybe you think it is too loose.

homepages.inf.ed.ac.uk/ldixon/cisa/files/potgieter-06-03-23.pdf

"In our research and technology, we are exploiting
self-awareness in order to engineer emergence in
complex adaptive systems. A complex adaptive system
learns from and adapts to its dynamically changing
environment. Such a system achieves self-awareness
by using an observation mechanism to observe its
own behavior and to update an internal model. This
self-awareness is then used to adapt its own behavior
in response to the ever-changing environment. This
engineering process "emerges" and is called emergent
engineering."



This paper is a pretty bad example. He says the following ...

"... A complex adaptive system learns from and adapts to its
dynamically changing environment. Such a system achieves self-awareness
by using an observation mechanism to observe its own behavior and to
update an internal model. This self-awareness is then used to adapt its
own behavior in response to the ever-changing environment. This
engineering process "emerges" and is called emergent engineering.
All complex adaptive systems in nature employ automatic emergent
engineering...".

I have no problems with the term emergence, but using self-aware here
again just muddies the water like Melanie.

Interestingly, Potgeiter's only 2 references in the paper are to
Holland, who does NOT use the term self-aware to refer to his "complex
adaptive systems", and to Minsky's A- and B-brains from SOM that I
mentioned last time.

As I already noted last time, B-brains are a good exmaple of using a
pattern recognition device [B-brain] to "monitor" the operation of
another device [A-brain], but this is hardly self-awareness anything
like human level.

Very muddy waters here.



This is turning into another one of these stupid discussions about
semantics and dueling definitions.

Funny, I checked back in SOM and see that Minsky says on page 59 ....
"To the extent that the B-brain knows what is happening in A, the
entire system could be considered to be partly "self-aware" ...", so I
guess he is in some agreement with Potgeiter. Depending on the
fuzziness of "partly". A tiny bit? A lot?

However, from what I've seen, I wonder that Minsky would accept this
statement of Potgeiter's ... "... The A-Brain and B-Brain engineer
emergence 'automatically' ...", as Minsky seems not to like the idea of
emergence very well. I may be wrong.

.

Relevant Pages

  • Re: Paper by ~MM on distributed self-awareness
    ... consciousness. ... some way connected to "self-awareness", but this is really just the ... As Melanie said, 'they were more likely to get et ... So originally "emergence" was defined as irreducible, ...
    (comp.ai.philosophy)
  • Re: The Emergence of Birthday
    ... author referred to as emergence. ... I am in agreement with Wilkins on emergence. ... or name what chemistry and other physics does in a roughly 30km shell ... Reflection such as this is key to building complex adaptive systems, complex systems which are often distributed and which have the capability to learn and adapt from experience. ...
    (talk.origins)