On 7/7/21, Mike Archbold <[email protected]> wrote:
> I'd benefit from a few paragraphs of appeals to intuition before
> diving into the formalisms, although I know the style for this type of
> paper is a sort of compactness.

In fact my theory does not make use of category theory in any
significant way...  not even the Curry-Howard isomorphism (as
I realize just now)...

The most tangible result is the symmetric NN and it only
requires that the logic conjunction is commutative.  Another
"tangible result" is that BERT can be seen as a logic, and
BERT is also "tangible" and it sort of provides evidence that
the logic-based approach seems not too far from realization....

It took me a lot of effort to establish a connection from AGI
to categorical logic...  this is like describing AGI in a new
language, but it's not simply a translation with a "dictionary".
The important thing is that categories also describe a lot
of things familiar to mathematicians:  groups, rings, fields,
vector spaces, polynomials, algebraic curves and surfaces,
etc.  This is their "daily bread".  Once the connection is
made there is no telling what useful things other mathematicians
may discover....

Also, the categorical abstraction allows me to see more
clearly what is the entire mathematical structure of logic....
if we don't have this picture we may think logic is a very
complicated machinery and we may never eliminate the
feeling that something is still "missing" in our considerations....

(I will add something like this to the introduction)

------------------------------------------
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/Tb5526c8a9151713b-M39938e0908a82cd4ad0f69ff
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Reply via email to