No reason AGI's shouldn't have any human capability - but your question
suggests a typical AGI-er naivete about the *evolution* of capabilities.
Being able to a) understand a dramatic video and b) assess it on any
aesthetic criteria, involves massively sophisticated, evolved capabilities -
a long, long way off an agent or human being able to simply navigate and
manipulate its environment - wh. are the natural starting points. There is
a loose *order* of evolution for capabilities, wh. we are going to have to
discover.
"No representation without navigation and manipulation" strikes me as a
reasonable, essential platform for *practical * AGI - for a *serious*
field of AGI, that wants to set out about reproducing/evolving human/animal
faculties in a reasonable order. First you gotta move and perceive - only
much later can you engage in reflection about what you're doing, let alone
get into learning incredibly evolved maths to analyse your actions.
What we have now - the pure fantasy that machines can leap to human-level
intelligence (incl. maths and aesthetic appreciation) - in a couple of
bounds - that there is no natural order of evolution of capabilities - is,
to use Sergio's word, absolutely "preposterous". A Don Quixote joke of pure
dreamers.
P.S. Manipulation of one's environment clearly evolves in very elaborate
stages too - and comes after locomotion.
P.P.S. "The order of evolution of faculties" (or some similar phrase) is
a vital concept for AGI . I can't recall coming across any equivalent - is
there one?
-----Original Message-----
From: Matt Mahoney
Sent: Wednesday, September 12, 2012 1:43 PM
To: AGI
Subject: Re: [agi] Discovering physical dimensions
On Tue, Sep 11, 2012 at 9:31 PM, ARAKAWA Naoya <[email protected]>
wrote:
On 2012/09/12, at 0:19, Mike Tintner <[email protected]> wrote:
AT:Well, let's just say that one of my working hypotheses is that
"seeing" may be too little for AGI.
Well, the question is - why does an agent need sight anyway?
We don't understand (*are* there any theories?) at what point an agent
needs sight.
An agent doesn't need sight (until a particular task requires it).
There are many blind people conducting independent lives.
When you do a Google search for "funny videos", the results should
ideally be ranked by a program that judges the humor content of
videos. This is something that humans can do. Is there any reason that
AGI should not have this capability?
-- Matt Mahoney, [email protected]
-------------------------------------------
AGI
Archives: https://www.listbox.com/member/archive/303/=now
RSS Feed: https://www.listbox.com/member/archive/rss/303/6952829-59a2eca5
Modify Your Subscription:
https://www.listbox.com/member/?&;
Powered by Listbox: http://www.listbox.com
-------------------------------------------
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
