Mike Dougherty wrote:
On 10/4/07, Richard Loosemore <[EMAIL PROTECTED]> wrote:
Do it then. You can start with interesting=cyclic.
should GoL gliders be considered cyclic?
I personally think the candidate-AGI that finds a glider to be similar
to a local state of cells from N iterations earlier to be particularly
astute. (assuming this observation is learned rather than hard-coded
by the developer)
Human 'players' of GoL will stop a run after reaching a stable cycle
because it is no longer interesting. The collection of cells
comprising a glider stops being interesting when we predict that it
will never 'hit' anything. Some of the seeds that ship with popular
GoL implementations are absolutely amazing. I'm sure after I
understand their nature the novelty will wear off. :)
I've been thinking about GoL, pattern recognition, vision, concept
representation. If have some ideas that I'd like to experiment with,
but I'm not really sure yet how to express them let alone implement a
test.
All understood. Remember, though, that the original reason for talking
about GoL was the question: Can there ever be a scientific theory that
predicts all the "interesting creatures" given only the rules?
The question of getting something to recognize the existence of the
patterns is a good testbed, for sure.
Richard Loosemore
-----
This list is sponsored by AGIRI: http://www.agiri.org/email
To unsubscribe or change your options, please go to:
http://v2.listbox.com/member/?member_id=8660244&id_secret=50175211-af906e
