From: Phil Goetz (firstname.lastname@example.org)
Date: Mon Aug 22 2005 - 10:37:06 MDT
--- Michael Wilson <email@example.com> wrote:
> Phil Goetz wrote:
> > I thought it was a pretty good list, except that it
> > left out control theory and signal processing.
> I agree with you in principle, but these things look less
> important than they used to. It's funny, once upon a time
> I used to be in favour of explicitly specifying
> peripheral complexity like that and leaving the core,
> important design complexity to 'emergence'.
That's not at all what I meant. I'm trying to figure out
the connection between control theory and signal
and peripheral vs. core complexity, and not getting it.
Look at Chris Eliasmith's book, "Neural engineering".
This is an excellent start on constructing modular systems
out of neural networks. You can begin to see how you
might solve the 100-step problem, when you see how to
construct neural networks to implement each of those 100
steps, where each step is a signal processing step such as
performing a Fourier transform, or mapping deltas into
(In its current incarnation, Neural Engineering
is at odds with dynamic systems theory approaches,
because of its modular nature.)
Control theory by itself is not going to make an AGI, but
control theory techniques may well be applied to the cyclic
or chaotic attractors stored in memory in order to
stabilize them, to make use of them as a pattern generator
to structure movement OR "thought".
> The only paper I can think of off hand that makes a
> semi-reasonable case for dynamic systems theory as the
> basis for AGI is 'Dynamics and Cognition' by Timothy van
> Gelder. And he didn't have any good reasons why one would
> /want/ to use that as a basis for AGI design, only some
> vague arguments as to why it should be possible and how
> it might be useful for analysing the brain.
Now I don't understand the phrase, "the basis for AGI".
Nothing from any of these communities would be "the basis".
You may have many necessary components, none of which
are "the basis".
We have evidence that, at least in some cases,
brains use attractors, possibly chaotic ones,
as memory elements. This includes Freeman's work
studying the patterns present in olfactory cortex
on exposure to and while learning different smells.
It also includes the plots of human cyclical movements
as described in Kelso's work. We have good theories as
to why this should be the case, including the properties
of switching dynamics (see my 1998 paper, Goetz and
Walters, "The dynamics of recurrent behavior networks,"
_Adaptive Behavior_ 6(2): 247-283), and the necessity of
having "fuzzy" (but not merely fuzzy-logic-like)
representations AS WELL AS fuzzy operators to work on those
representations, in order to have flexible categorization,
analogical reasoning, metaphor, creativity, etc. In
addition, the time aspect of representation and action is
very poorly accounted-for by non-dynamical approaches.
One could fairly say it is ignored entirely, and that
experiments are structured to sidestep it. There is a bit
on this in Jeff Hawkins' new book /On Intelligence/, IIRC.
- Phil Goetz
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:52 MDT