Shane,

> Yes, this is the basic idea.  Some structures and dynamics will tend
> to categorize and cluster information, some will tend to recognize
> temporal patterns, some might tend to store information, some will
> be mixtures of these things.  Exactly what I'll find is hard to know
> in advance.  As you say this isn't a powerful "brain", but it might
> serve as a set of useful building blocks to construct more powerful
> general intelligence from.

Agreed

> Your approach seem to be to skip this first step by careful design.
> The danger is that it's hard and you might mess it up.  Some of the
> needed dynamics could be very subtle and easily missed.  Also it's
> not clear just how "small" the fundamental units of representation
> have to be in order to be flexible and dynamic enough.

Well, my units of representation are mostly objects containing pairs of
numbers.

Not significantly more complex than formal neurons, and much simpler than
biological neurons.

But I have some units of representation that represent simple operations
like AND, NOT or the S or K combinators.  These latter are a little more
complex than formal neurons, but still much simpler than biological neurons.

My dynamical equations are definitely not more complex than, say, the
Hodgkin-Huxley equations governing the dynamics of real neurons.  (But
they're quite different.)  Anyway, they're relatively small algebraic
formulas.

One point is that Novamente (unlike Webmind) doesn't consist of large units
which in turn consist of entirely of local agents doing local computations.
In Novamente, the large units also contain some global computations.

> The danger with my approach is that my theory might be too weak.
> This could leave me with a search space that is too big to effectively
> search.  Also, if it turns out that my fundamental units of
> representation are smaller than what is really needed, I might produce
> very inefficient solutions.

I don't think that "smaller" is the right criterion to be looking at....
The key point is the type of dynamics that one is using.  Novamente's units
of representation are pretty small, but the dynamics are quite different
from those in neural net models.  The big question for you will be how you
restrict the space of possible dynamics to search over.

-- Ben G

-------
To unsubscribe, change your address, or temporarily deactivate your subscription, 
please go to http://v2.listbox.com/member/[EMAIL PROTECTED]

Reply via email to