[Fis] Fis Digest, Vol 105, Issue 12 Human Wisdom vs Meta-Intelligence. Dreaming

Karl Javorszky karl.javorszky at gmail.com
Tue Nov 7 13:09:22 CET 2023


>
> Dreams
>
2023 11 07

To have dreams, the system needs at least two properties


1. To have a reordering phase, during which such elements come to be
associated with each other which are not associated with each other during
the production phase ;
2. The associations during the reordering phase do not follow the rules
valid for associations that appear during the production phase.

Example:
The movements of the chess figures at the end of the game during the
reordering into the standard starting positions.

Applicability :
The elements of the Tautomat move according to periodical changes (tidal,
daily, yearly). These periodicities yield the "starting positions in a game
of chess". Inanimate objects follow the changes. Deviations "expected place
vs observed places" are at minimum. The elements are attached to their
places by the extent of the liens of the periodicities.

The Tautomat is now instructed to look for the least attached of the
elements and see whether a 4th variant of reordering is achievable from
those and other, and at what cost. It finds one of such and then more of
such, where the liens built on non-periodic associations can compete with
the liens built on periodic associations. It can do such up to a limit, as
long as it doesn't contradict the requirements of the perioic changes. The
Tautomat has to reset and the human has to sleep in order to realign to the
orders that are dictated by the periodic changes. (death as the ultimate
realignment).

Yes, if machines are instructed to behave otherwise than expected at first
glance (to pick other than the apparently best alternative, eg for
individuality or curiosity), then it will be necessary to keep count of the
deviations to the most common alternatives. The reset has the form of sleep
in higher organisms. If we instruct the machine to log the reset dump, we
will gain entry to its dreams.

Karl
------------ pr�xima parte ------------
Se ha borrado un adjunto en formato HTML...
URL: <http://listas.unizar.es/pipermail/fis/attachments/20231107/8d986f12/attachment.html>


More information about the Fis mailing list