Matt,
Jeez, massive question :).
Let me 1st partly dodge it, by giving you an example of the difficulty of
understanding, say, "over", both in NLP terms and ultimately (because it
will be the same more or less) in practical object recognition/movement
terms - because I suspect none of you have done what I told you, (naughty)
& looked at Lakoff.
You will note the very different physical movements or positionings involved
in:
The painting is over the mantle
The plane flew over the hill
Sam walked over the hill
Sam lives over the hill
The wall fell over
Sam turned the page over
She spread the cloth over the table.
The guards stood all over th ehill
Look over my page
He went over the horizon
The line stretches over the yard
The board is over the hole
[not to mention]
The play is over
There are over a hundred
Do it over, but don't overdo it.
& there are many more.
See Lakoff for schema illustrations. Nearly all involve very different
trajectories, physical relationships.
That is why I'm confident that no program can handle that, but yes, Mark, I
was putting forward a new idea (certainly to me) in "the orientation
framework" - and doing no more than presenting a reasoned, but pretty
ill-informed hypothesis. (And that is what I think this forum is for. And I
will be delighted if you, or anyone else, will correct my
overgeneralisations and errors).
Now a brief, rushed but, I suspect, massive, and new answer to your
question - that I think, takes us, philosophically, way beyond the concept
of "grounding", which a lot of people are currently using for
"understanding."
To "understand" is to "REALISE" what [on earth, or in the [real] world] is
being talked about. It is, in principle, and often in practice, to be able
to go into the real world and point to the real objects/actions being
referred to, (or realise that they are unreal/fantastic). So in terms of
understanding a statement containing how something is over something else,
it is to be able to go and point to the relevant objects in a scene, or, if
possible, to recreate the physical events or relationship..
I believe that is actually how we *do* understand, how the brain does work,
how a GI *must* work - &, if correct, it automatically moves us beyond
"virtual AGI". I shall hopefully return to this concept on further
occasions - I believe it has enormous ramifications. There are many, many
qualifications to be made, which I won't attempt now, nevertheless the basic
principle holds - and will hold for the psychology of how humans understand
or *don't* understand or get confused.
IOW not only must an AGI or any GI be "embodied" it must also be directly &
indirectly "embedded" in the world.
("Grounding" is being currently interpreted in practice almost entirely from
the embodied or agent's side - as referring to what goes on *inside* the
agent's mind. "Realisation" involves complementarily defining intelligence
from the out-side of its ability to deal with the environment/real world
being-referred-to. BIG difference. Like between just using nature/heredity,
OTOH, and, OTOH, also using nurture/environment to explain behaviour).
I hope you realise what I've been saying :).
Matt:
Mike, your argument would be on firmer ground if you could distinguish
between when a computer "understands" something and when it just reacts as
if it understands. What is the test? Otherwise, you could always claim
that a machine doesn't understand anything because only humans can do
that.
-- Matt Mahoney, [EMAIL PROTECTED]
--- On Thu, 9/11/08, Mike Tintner <[EMAIL PROTECTED]> wrote:
From: Mike Tintner <[EMAIL PROTECTED]>
Subject: Re: [agi] Artificial humor
To: [email protected]
Date: Thursday, September 11, 2008, 1:31 PM
Jiri,
Clearly a limited 3d functionality is possible for a
program such as you
describe - as for SHRDLU. But what we're surely
concerned with here is
generality. So fine start with a restricted world of say
different kinds of
kid's blocks and similar. But then the program must be
able to tell what is
"in" what or outside, what is behind/over etc. -
and also what is moving
towards or away from an object, ( it surely should be a
"mobile" program) -
and be able to move objects. My assumption is that even a
relatively simple
such general program wouldn't work - (I obviously
haven't thought about this
in any detail). It would be interesting to have the details
about how SHRDLU
broke down.
Also - re BillK's useful intro. of DARPA - do those
vehicles work by GPS?
> Mike,
>
> Imagine a simple 3D scene with 2 different-size
spheres. A simple
> program allows you to change positions of the spheres
and it can
> answer question "Is the smaller sphere inside the
bigger sphere?"
> [Yes|Partly|No]. I can write such program in no time.
Sure, it's
> extremely simple, but it deals with 3D, it
demonstrates certain level
> of 3D understanding without embodyment and there is no
need to pass
> the orientation parameter to the query function. Note
that the
> orientation is just a parameter. It Doesn't
represent a "body" and it
> can be added. Of course understanding all the
real-world 3D concepts
> would take a lot more code and data than when playing
with 3D
> toy-worlds, but in principle, it's possible to
understand 3D without
> having a body.
>
> Jiri
>
> On Thu, Sep 11, 2008 at 11:24 AM, Mike Tintner
<[EMAIL PROTECTED]>
> wrote:
>> Jiri,
>>
>> Quick answer because in rush. Notice your
"if" ... Which programs
>> actually
>> do understand any *general* concepts of
orientation? SHRDLU I will gladly
>> bet, didn't...and neither do any others.
>>
>> The v. word "orientation" indicates the
reality that every picture has a
>> point of view, and refers to an observer. And
there is no physical way
>> around that.
>>
>> You have been seduced by an illusion - the
illusion of the flat, printed
>> page, existing in a timeless space. And you have
accepted implicitly that
>> there really is such a world -
"flatland" - where geometry and
>> geometrical
>> operations take place, utterly independent of you
the viewer and
>> puppeteer,
>> and the solid world of real objects to which they
refer. It demonstrably
>> isn't true.
>>
>> Remove your eyes from the page and walk around in
the world - your room,
>> say. Hey, it's not flat...and neither are any
of the objects in it.
>> Triangular objects in the world are different from
triangles on the page,
>> fundamentally different.
>>
>> But it is so difficult to shed yourself of this
illusion. You need to
>> look
>> at the history of culture and realise that the
imposition on the world/
>> environment of first geometrical figures, and
then, more than a thousand
>> years later, the fixed point of view and
projective geometry, were -
>> and
>> remain - a SUPREME TRIUMPH OF THE HUMAN
IMAGINATION. They don't exist,
>> Jiri. They're just one of many possible
frameworks (albeit v useful) to
>> impose on the physical world. Nomadic tribes
couldn't conceive of squares
>> and enclosed spaces. Future generations will
invent new frameworks.
>>
>> Simple example of how persuasive the illusion is.
I didn't understand
>> until
>> yesterday what the "introduction of a fixed
point of view" really meant -
>> it
>> was that word "fixed". What was the big
deal? I couldn't understand.
>> Isn't
>> it a fact of life, almost?
>>
>> Then it clicked. Your natural POV is
"mobile" - your head/eyes keep
>> moving -
>> even when reading. It is an artificial invention
to posit a fixed POV.
>> And
>> the geometric POV is doubly artificial, because it
is "one-eyed", no?,
>> not
>> stereoscopic. But once you get used to reading
pages/screens you come to
>> assume that an artificial fixed POV is *natural*.
>>
>> [Stan Franklin was interested in a speculative
paper suggesting that the
>> evolutionary brain's "stabilisation of
vision", (a software triumph
>> because
>> organisms are so mobile) may have led to the
development of
>> consciousness).
>>
>> You have to understand the difference between 1)
the page, or medium,
>> and
>> 2) the real world it depicts, and 3) you, the
observer, reading/looking
>> at
>> the page. Your idea of AGI is just one big page
[or screen] that
>> apparently
>> exists in splendid self-contained isolation.
>>
>> It's an illusion, and it just doesn't
*work* vis-a-vis programs. Do you
>> want to cling to "excessive optimism"
and a simple POV or do you want to
>> try
>> and grasp the admittedly complicated & more
sophisticated reality?
>> .
>>
>> Jiri: If you talk to a program about changing 3D
scene and the program
>> then
>>>
>>> correctly answers questions about [basic]
spatial relationships
>>> between the objects then I would say it
understands 3D. Of course the
>>> program needs to work with a queriable 3D
representation but it
>>> doesn't need a "body". I mean it
doesn't need to be a real-world
>>> robot, it doesn't need to associate
"self" with any particular 3D
>>> object (real-world or simulated) and it
doesn't need to be self-aware.
>>> It just needs to be the 3D-scene-aware and the
scene may contain just
>>> a few basic 3D objects (e.g. the Shrdlu
stuff).
>>>
>>
>>
>>
>>
>> -------------------------------------------
>> 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/?&;
>> 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/
> 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/
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/
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/
Modify Your Subscription:
https://www.listbox.com/member/?member_id=8660244&id_secret=111637683-c8fa51
Powered by Listbox: http://www.listbox.com
