You're trying for an extreme position. I didn't and don't argue that you
can't use maths in a program, (although I severely doubt your sweeping
claims about AI and programs not being able to function without maths).
I am merely focussing on the AGI *dimension* of a truly intelligent agent.
Total intelligence must embrace both a creative (AGI) dimension AND a
rational routine (narrow AI) dimension.
It is for that AGI dimension that maths is self-evidently useless. There is
nothing in maths that is creative or generative. Maths consists of
*analyses* of existing/"old" forms and numbers - not ideas for generating
*new* forms and patterns and numbers. *Mathematicians* are creative - but
their creativity does not come from maths. It isn't existing formulae etc
that generate new formulae, or existing patterns that generate new
patterns. (Contrast patchworks which are endlessly generative). It is the
mathematician's iconic imagination.
One proof is in the pudding of how mathematically inclined AGI-ers like Legg
and Hutter, actually define intelligence. They *completely ignore* the
creative aspect - completely ignore what scientific psychology tells us -
that intelligence consists of both creative {fluid,divergent, "wild",
"unprogrammed"/"unstructured" [per Simon] intelligence and rational
[crystallised, convergent, "tame", "programmed", "structured" intelligence.
They make no formal distinction between AGI and narrow AI - because maths
has no means of formalising creativity.
So I've begun to offer a formal approach to how we can
computationally/robotically be creative - an embodied iconic approach -
which happens to coincide with the science of both mirror neurons and
embodied cognitive science.
Frankly, anyone who doesn't look at that whole realm and kind of approach is
just closed-minded and blinkered - a mathematical Luddite - who won't have
produced anything that is really AGI, and never will.
P.S. Creativity - object recognition - and conceptualisation - are all one
and the same. The problem of object recognition and conceptualisation is
largely to deal with a creative world which continually presents us with
new, creative kinds of objects and object actions. The problem of creativity
is not just to react to, but to proactively produce new kinds of objects and
actions.
-----Original Message-----
From: Ben Goertzel
Sent: Sunday, January 06, 2013 4:23 PM
To: AGI
Subject: Re: [agi] Why Logic & Maths Have Sweet FA to do with Real world
reasoning
On Sun, Jan 6, 2013 at 5:36 AM, Mike Tintner <[email protected]>
wrote:
What’s the point of this except to avoid having to think about a new idea?
The point of that post was to clearly state that
"Math is irrelevant to AGI"
implies
"All known or currently envisioned digital, analog or quantum
computers are irrelevant to AGI"
since any program on any of these computers has a mathematical
formulation; and furthermore math is the tool used to design these
computers and the operating systems languages that operate on them,
etc.
New ideas regarding how to program computers to yield AGI would be
interesting to hear. But these ideas will inevitably have mathematical
formulations, else they can't be implemented on any known or
envisioned computers. You may choose to describe them
non-mathematically due to your own specific taste and background; but
this may then make your ideas harder for those of us with
scientific/technical background to understand...
.. ben g
If an “iconic program” used logic and/or maths, they would only be parts,
not the whole program.
The point of doing AGI is to solve the unsolved problems by identifying
new technologies – not to use old technologies that show no signs of
working. Your loyalty is to the old technologies. Old technologies never
solve new creative technology problems. That’s just a wrong idea. We didn’t
get to jets by adapting the propellor, or to neural nets by adjusting
linear
programs.
From: Ben Goertzel
Sent: Sunday, January 06, 2013 1:27 AM
To: AGI
Subject: Re: [agi] Why Logic & Maths Have Sweet FA to do with Real world
reasoning
But if the robot is controlled by a computer program, that program is
still embodying some logico-mathematical formula, at any particular point
in
time...
-- Ben G
On Sat, Jan 5, 2013 at 8:01 PM, Mike Tintner <[email protected]>
wrote:
P.S. What I should have spelled out is that the moment you connect an
icon to a body -
for example, an icon of a hand to a physical hand
**the icon automatically becomes a fluid schema**
because a hand – like the body - is in effect a fluid schema itself.
What can an agent do with a hand? Nobody can state a definite bounded set
or
frame of hand movements. A hand is capable of virtually infinite hand
movements.A fluid range of movements with many yet to be discovered and
many
continually being invented. So a hand icon [when connected to a
body/hand]
ipso facto refers to a potentially infinite range of hand
movements/shapes
etc.
No one shall enter into the kingdom of AGI who is not a roboticist.
AGI | Archives | Modify Your Subscription
--
Ben Goertzel, PhD
http://goertzel.org
"My humanity is a constant self-overcoming" -- Friedrich Nietzsche
AGI | Archives | Modify Your Subscription
AGI | Archives | Modify Your Subscription
--
Ben Goertzel, PhD
http://goertzel.org
"My humanity is a constant self-overcoming" -- Friedrich Nietzsche
-------------------------------------------
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-f452e424
Modify Your Subscription:
https://www.listbox.com/member/?member_id=21088071&id_secret=21088071-58d57657
Powered by Listbox: http://www.listbox.com
