AT: there is no reason to believe that a self-organizing system would
reach the same solution through a "staged", staggered input (eg. 1.
static images 2. obscured static 3. moving 4. obscured moving 5.
morphed moving objects etc).
This is a stimulating comment.
Workers in this field do really need to take time to hold up the
philosophical foundations of their projects to the light before they get too
involved.
Kuipers talks of the "hose" of experience. Well, the arguably key thing to
realise here - esp when thinking about bootstrapping projects - is that the
sensors, motor ability, perceptual and reflective ability, and needs and
life-of-activities of any agent are bound to be interrelated. The general
principle must almost certainly be that they evolve - become more and more
complex - together. Systems will evolve in a unified way. Simple organisms
for example don't need to see much. It's only complex organisms like us that
to need to see the kind of detail that we do.
That's just one reason why you *have* to start small and simple - your
chances of achieving an evolutionary "systemic balance" otherwise are
non-existent.
-------------------------------------------
AGI
Archives: https://www.listbox.com/member/archive/303/=now
RSS Feed: https://www.listbox.com/member/archive/rss/303/21088071-c97d2393
Modify Your Subscription:
https://www.listbox.com/member/?member_id=21088071&id_secret=21088071-2484a968
Powered by Listbox: http://www.listbox.com
