From: Eugene Leitl (eugene.leitl@lrz.uni-muenchen.de)
Date: Sat May 15 1999 - 19:35:08 MDT
Anders Sandberg writes:
> Distributed genetic algorithms to try to find good solutions to
> problems? But the set of interesting problems with easily calculated
> fitness functions may be small; most of the really interesting
> problems (like AI or making good search agents) have rather diffuse
> fitness.
Why not utilizing open-ended evolution? It would be rather difficult
to code, though.
This archive was generated by hypermail 2.1.5 : Fri Nov 01 2002 - 15:03:44 MST