Re: determinism, freewill, chaos, and circular causality

"feedbackdroid" <feedbackdroid@xxxxxxxxx> wrote:
Curt Welch wrote:



I've never had any issues understanding that for one, there are events
which we can't know the cause of (QM and other events which happen at
the limits of our ability to measure). Whether they have a cause or
not is something we will probably never know - and no matter what
science uncovers in the future to explain the cause, there's likely
always going to be some lower level event which will replace it which
we won't know the cause of. And at the same time, I've also had no
issue with understanding how these events can be amplified through
iteration and feedback effects to create what is generally know as
chaotic systems. All this seems fairly obvious to me. It's the source
of all noise in the universe.

What I take issue with however and what strikes me as invalid, is to
believe that these types of effects have something important to do with
the operation of the brain or intelligence. On the contrary, I see the
entire purpose of the brain as being a system which must find, and
amplify, all the non chaotic effects of the universe. Brains are in
effect, noise filters designed to remove all the noise which the
universe is full of. They are anti-chaotic devices if anything.



The first thing is that the events referred to occur at a level far far
"above" QM effects, and likely have very little to do with QM. Every
physical system in the universe can ultimately be reduced to atoms and
quarks and QM whatevers, but these chaotic effects are occuring at
"macro-levels" of interaction between the units. That's the first
thing. The 2nd thing is that the brain has all of the same fundamental
aspects as every other system that exhibits chaotic behavior. Thirdly,
from Briggs+Peat again ...

pg 145: "... Biological systems remain stable by damping most small
effects except in those areas of behavior where a high degree of
flexibility and creativity is required. Here the system remains highly
sensitive to to influx, close to a state of chaos...".

Likewise, a chaotic system can have relatively stable modes
[attractors], in which case its sensitivity to noise is minimal, but
when given the correct input it can move to another attractor.

Of course. Which is exactly how the brain takes the noise out of the
equation - by creating attractors.

Most of what you quoted above I agree with - except that last part
where they seem to be justifying noise as a source of creativity. It's
even worded in way to imply that maybe, without a universe full of
noise, there would be no creativity. I don't agree with that at all.
I believe creativity is nothing more than the result of a system which
has the power to search a design space for useful solutions.



In a highly complex system, with many interacting units and feedback
loops, there are potentially an astronomical number of different
operational states [attractors] that the system can go between.

Sure. Complex systems can have a lot of states. It's that obvious without
the need to bring in feedback, attractors, or chaos?

Electrical logic gates and digital memory systems are attractors designed
to attract their state to one value or another - yet we don't need chaos
theory or the idea of "attractors" to understand their operation. A
digital memory with only 64 bits (binary attractors) of memory has 2^64
different states. How is any of this new? How does talking about this in
terms of chaos theory and using words like attractors help us to understand
that even small numbers of logic gates can create an astronomically huge
state space?

It makes no difference if
the search is random and chaotic and influenced by unpredictable low
level noise sources or whether it's totally deterministic.



Actually, this is an important point with chaotic systems, as once near
a "bifurcation point", then noise can drive operation in an
unpredictable direction from there.

Just like noise will drive a logic gate into an unpredictable direction if
it's near the border between two attractors. The point is that
unpredictable behavior is not intelligent behavior. Intelligent behavior
happens when you eliminate unpredictable behavior and replace it with
predictable behavior.

Human behavior should not be looked at as unpredictable. It's simply
complex. There's a big difference. If intelligent behavior was in fact
unpredictable and chaotic, we would all be dead.

The end result is
just the fact that "good" options are found. There are plenty of
examples of genetic algorithm programs which are 100% deterministic but
yet have no problem being extremely creative (creating things that no
human was able to create). Creativity is just the ability of the
system to recognize a good solution when it sees it and to take
advantage of all the good solutions it finds.



... and to get into such a state in the first place.



I just keeping seeing all these attempts to link chaos and QM and other
hard to understand effects with a desperate attempt to explain the



As mentioned, I think there is nothing whatsoever to do with linking QM
and chaos - in general. Maybe a lot of people have trouble because they
believe these are the same thing.

That's true. Chaotic behavior is easily created in a very predictable
deterministic environment with feedback - just like PRNGs act as chaos
machines created with the simple application of feedback. It's just
strikes me that the people that are attracted to chaos as an answer to the
complexity of human behavior also tend the be the same lot that are
attracted to QM and other low level and unpredictable (or hard to predict)
effects.

I've got nothing against the people studying chaos. It's a very
interesting field with lots of application to us understanding the nature
of the universe (which is mostly chaotic). I just don't think it is going
to to do much to help us understand AI or the human brain.

complexity of human behavior by people who have failed to appreciate
(or find) the simplicity behind human behavior. It's as if they
believe that it can't be the result of something simple or else someone
would have uncovered the answer - so they turn to everything complex in
hopes of finding the answer. I just don't agree with that because I'm
sure that intelligence is simple. I don't think it could work if it
was complex.



Of course, my conception is exactly the opposite. IOW, there is a
reason it takes 100B neurons and 100T synapses to produce human-level
intelligence. Ape brains, for instance, have maybe 1/10 the number of
neurons as humans, and their intelligence is a lot less for it. So, a
fundamental question is, why can you get such a tremendous increase in
intelligence, especially in a "qualitative" rather than just a
quantitative way, by increasing the #units by that amount?

I think the big win for humans happens with our use of language. I
strongly suspect that the configuration of our brains has far more to do
with our language skills than the size of our brains. I suspect that an
ape-size brain, if correctly re-configured, would give apes far stronger
language skills than they currently have, and put them close in
intelligence to humans. I think comparing size alone misses the most
important aspect of why we have such strong language skills and comparing
the size of an ape brain to the size of a human brain won't tell us
anything useful about why we have stronger abstract symbol manipulation
skills than apes.

--
Curt Welch http://CurtWelch.Com/
curt@xxxxxxxx http://NewsReader.Com/
.

Relevant Pages

  • Re: determinism, freewill, chaos, and circular causality
    ... with many interacting units and feedback ... operational states [attractors] that the system can go between. ... to attract their state to one value or another - yet we don't need chaos ... there has to be a way to get the brain cells ...
    (comp.ai.philosophy)
  • Re: Artificial life likely in 3 to 10 years
    ... If you are just saying that a neuron is worth more than ... analog computing is usually a waste of time. ... would be wasted - due to the levels of noise in the brain. ... but the noise of inseparable part of brain functioning. ...
    (sci.bio.evolution)
  • Re: determinism, freewill, chaos, and circular causality
    ... It's the source of all noise in the universe. ... operation of the brain or intelligence. ... they seem to be justifying noise as a source of creativity. ...
    (comp.ai.philosophy)
  • Re: Deadly .22 rimfire
    ... # pieces will scramble the brain. ... The .22 subsonic was a favorite of the Mossad because the silenced Beretta ... would not make any noise. ... between speed and accuracy of a pistol round, it tends to be the opposite. ...
    (rec.guns)
  • Re: determinism, freewill, chaos, and circular causality
    ... It's the source of all noise in the universe. ... operation of the brain or intelligence. ... they seem to be justifying noise as a source of creativity. ...
    (comp.ai.philosophy)