I have often thought this is the best architecture for an AGI (or brain for a
mind). But those who have tried to implement a massively parallel system
modelling neurons obviously hasn't been successful, at least not yet. So as
programmers, we perform tricks and come up with algorithms to try and take
shortcuts.
Instead of modelling one metaphorical ant, would the system evolve and be self-
organized if we modelled a group of ants (an anthill?) instead? This would be
more computationally efficient, but we would not end up with the level of
intelligence we expect as it self-organizes.
cheers,
Simon
> An interesting approach to AI by modeling ant colony behavior..
>
> http://www.openp2p.com/pub/a/p2p/2003/02/21/bonabeau.html
-------
To unsubscribe, change your address, or temporarily deactivate your subscription,
please go to http://v2.listbox.com/member/[EMAIL PROTECTED]
