On Tue, Aug 5, 2008 at 12:48 AM, Ben Goertzel <[EMAIL PROTECTED]> wrote:
> The problem is that writing stories in a formal language, with enough nuance
> and volume to really contain the needed commonsense info, would require a
> Cyc-scale effort at formalized story entry.  While possible in principle,
> this seems infeasible...

AGI capabilities can IMO be tested with much less data. When the data
representation, grounding, & general problem solving are proved to be
well working, then I'm planning to experiment with NL-FL conversions.
If an AGI cannot handle a formal language input (=the easiest input to
parse) then it IMO shouldn't mess with [much] harder-to-parse inputs
(e.g. 3D).

> Whereas, interactions in virtual worlds and games are happening already,
> round the clock, right now ... just waiting for AI's to get involved w/ them
> and learn from them...

When do you think Novamente will be ready to "go out" and effectively
learn from (/interract with) environments not fully controlled by the
dev team?

Jiri


-------------------------------------------
agi
Archives: https://www.listbox.com/member/archive/303/=now
RSS Feed: https://www.listbox.com/member/archive/rss/303/
Modify Your Subscription: 
https://www.listbox.com/member/?member_id=8660244&id_secret=108809214-a0d121
Powered by Listbox: http://www.listbox.com

Reply via email to