From: Ben Goertzel (ben@goertzel.org)
Date: Mon May 08 2006 - 06:23:39 MDT
> No system can be designed which is able to optimise it's own source code.
> I would have thought that Godel's Incompleteness Theorem would have made
> that abundantly clear.
Godel's Theorem places limitations on self-understanding,
self-optimization and goal-directed self-modification, but it
certainly does not prevent these things
> The solution is actually rather simple -
>
> You use TWO or more separate systems (with different code structures) to
> modify each other.
Pardon me, but this is very silly.
Any limitation of this sort that applies to INDIVIDUAL self-modifying
AI system, obviously also applies to PAIRS or larger groups of
inter-modifying AI systems as well.
Otherwise, one could make a single AI system circumvent the
limitations by simulating internally a division into N inter-modifying
systems.
Heh... If only there were really a trivial solution like this ;-ppp
-- Ben G
This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:56 MDT