Anyone read about that electronic eye chip?

Seth Russell seth at robustai.net
Fri Apr 28 09:22:00 EST 2000


steve at tropheus.demon.co.uk wrote:

> >>    Too bad that FORTH is such an awful anachronistic
> >>    language. Nobody thinks in stacks ...  :-)
> >
> >Bull shit!  Everybody thinks in stacks.
>
> Some of the diagrams on your own web site would be very difficult
> to reduce to stacked thoughts. Objects and links, not stacks.

True, knowledge stored in the relationships between symbols
is not a stack, but such knowledge grows in stacks (heaps).
Also the active processes of awareness and change of that
knowledge (artificial symbolic thinking) are stacks.
I once wrote a system on those very diagrams reducing the active
processes to two 2 stacks:  attention and the program
location stack.

> Steve Wolstenholme
> Neural Network Applications for Windows
> http://www.tropheus.demon.co.uk

I can see why you reacted so strongly against stacks,
NNs certainly are not stacks.  I've always wanted to
see the more traditional symbolic thinking (using stacks)
combined with the power of NNs.   Have you ever thought
of applying NNs to document retrieval based on key
word clustering?

--
Seth Russell
Http://RobustAi.net/Ai/SymKnow.htm
Http://RobustAi.net/Ai/Conjecture.htm






More information about the Neur-sci mailing list