Pei,
First I carefully read your pdf which only talks re symbols:
"The meaning of a symbol is in its
relations with other symbols; the
truth-value of a statement indicates
how close it is to the evidence"
Below you talk about more - images/motor sequences etc - well how do they
integrate with your pdf diagrams re concept processing/ semantics? That's
rather important to set out isn't it, if you're diagramming?
And there seems to be a big hole in the centre of the below - and this is, I
think,
an important question : who or what is the part of the "software agent"
that does the "experience"-ing? Do you have a self - as we do - with a
unified sense of (and indeed location in) body and brain - that experiences
all its knowledge via emotions/ sensations? Whatever, you must define the
experiencing entity, and how it experiences, and you don't AFAICT,
Let me try and make this exaggeratedly simple to focus on:
how is NARS or any current attempt at an AGI different, in the final
analysis, from an electronic row of dominoes, that a human programmer sets
off ? I am not aware - which of course may be my ignorance - of any current
system that is any different from that, however massively complicated the
row of dominoes - or that could be said to "experience" anything..
Pei:
--- page 23 ---
... we do not mean that a word in a natural language gets its
meaning only by its relation with other words in the language, because human
experience is not limited to a language channel, but closely related to
sensation, perception, and action (Barsalou, 1999; Harnad, 1990). However,
the general principle is still applicable here, that is, a word gets its
meaning
by its experienced relations with the system's other experiential
components,
which may be words, perceptive images, motor sequences, and so on. In a
system
like this, the meaning of a word is much more complex than in a system
whose experience is limited to a language channel, but it does not rule out
the
latter case as a possible way for words (terms, symbols) to be meaningful.
For
example, a software agent can get all of its experience in this manner, and
we
cannot deny that it is genuine experience.
--- page 24 ---
The definition of meaning in NARS is similar to conceptual role semantics
and
semantic network (Harman, 1982; Kitchener, 1994; Quillian, 1968), where the
meaning of a concept (or word) is defined by the role it plays in a
conceptual
system (or a natural language). The difference between experience-grounded
semantics and those theories are:
• In NARS, the relations among terms are not definitional or linguistic, but
experienced relations that happen in the interaction between a system and
its environment, therefore they are dynamic and subjective in nature.
• In NARS, the relations between a term and others are concretely specified
by its extension and intension, consisting of inheritance relations, whose
meaning and properties are formally specified.
• In NARS, whenever a term is used, only part of its meaning is involved.
In other words, the "current meaning" of a term is not exactly its "general
meaning" in the long run.
Pei
On Feb 16, 2008 12:32 PM, Matt Mahoney <[EMAIL PROTECTED]> wrote:
--- Mike Tintner <[EMAIL PROTECTED]> wrote:
> PeI:> To test the power of "visual reasoning", here is a rough visual
> > explanation on two very different ways for "symbols" to get their
> > meaning:
> >
> > http://nars.wang.googlepages.com/wang.semantics-figure.pdf
> >
>
> Wow, I have to stop talking but this is really stimulating. Your
> paper/illustrations are v. useful as far as they go, but they are almost
> literally the tip of the iceberg. Your Experience-Grounded Semantics
> represents a flower/pot as a tree or net of attached symbols
>
> "plant - containing - blossom - round" etc
>
> Now can we please have the VAST attached clusters/ trees of images of
> flowers and pots that your brain has, and uses, to understand and
> process
> flowers/plants
That's called a semantic network. Words are associated with other words
that
appear near it in a large corpus of text, for example:
http://labs.google.com/sets?hl=en&q1=flower&q2=plant&q3=pot&q4=containing&q5=blossom&btn=Large+Set
I agree that non symbolic (e.g. visual) processing is important for
systems
with non-symbolic I/O.
-- Matt Mahoney, [EMAIL PROTECTED]
-------------------------------------------
agi
Archives: http://www.listbox.com/member/archive/303/=now
RSS Feed: http://www.listbox.com/member/archive/rss/303/
Modify Your Subscription: http://www.listbox.com/member/?&;
Powered by Listbox: http://www.listbox.com
-------------------------------------------
agi
Archives: http://www.listbox.com/member/archive/303/=now
RSS Feed: http://www.listbox.com/member/archive/rss/303/
Modify Your Subscription:
http://www.listbox.com/member/?&;
Powered by Listbox: http://www.listbox.com
--
No virus found in this incoming message.
Checked by AVG Free Edition.
Version: 7.5.516 / Virus Database: 269.20.6/1282 - Release Date: 2/15/2008
7:08 PM
-------------------------------------------
agi
Archives: http://www.listbox.com/member/archive/303/=now
RSS Feed: http://www.listbox.com/member/archive/rss/303/
Modify Your Subscription:
http://www.listbox.com/member/?member_id=8660244&id_secret=95818715-a78a9b
Powered by Listbox: http://www.listbox.com
