Re: NL, formal logic, machine language, etc.

On Tue, Aug 20, 2019, 11:00 AM ProjectParadigm-ICT-Program <
metadataportals@yahoo.com> wrote:

> Excellent stuff,
>
> We also need to look into currents of philosophy dealing with signs,
> concepts and objects (or as the philosophers call it "selves").
>

If I'm not wrong, this "current" is called Semiotics, of which Semantics is
one of its three branches, together with Syntax and Pragmatics (jfsowa.com
may have some references about this).

I'm currently trying to tackle this problems in a layered "semantic"
(quads) grammar metamodel and a series of functional rules (monadic functor
transforms) performing "inferences" / productions.

Best,
http://github.com/snxama/scrapbook


> Two currents: "pragmatist" and "idealist", and deal with whether concepts
> have another content conferred on to them by their use.
>
> Why is this important, because it boils down to whether natural language
> can be formalized using ontologies and logical frameworks in such a way
> that all semantic information is completely captured.
>
> Godel's Incompleteness Theorem suggests this is not possible in a form
> which guarantees completeness and consistency.
>
> Only consistency is attainable which implies there will always be
> information and thus knowledge that cannot be captured.
>
> The following captures some interesting things in an interdisciplinary way:
>
>
> https://www.amazon.com/Systems-Self-Organisation-Information-Interdisciplinary-Perspective/dp/1138609935
>
> From this point it is just a few steps to Buddhist logic, specifically,
> Sautrantika Madhyamaka Logic, which deals with Emptiness and is the system
> of logic closest to quantum physics and string theory, and shows us the
> limitations of conceptual frameworks and inherent problems with observation
> of the physical world through sensory input systems.
>
> The only piece of the puzzle missing is the psycholinguistic one.
>
> What is the most important takeaway in this all? Knowledge cannot be
> captured in full by formalized language, ontologies or conceptual
> frameworks or formal logic alone. In order for any of this to have any
> practical use, we must strive for consistency, and forgo completeness.
>
> Biologically inspired cognitive architectures strive to find and model
> after brain processes that deal with creating knowledge and obviously
> communicating or transmitting this, and are particularly interested in
> algorithms and processes that require limited expenditure of energy and
> computing power, thus not requiring massive amounts of data and deep
> learning.
>
> Since completeness is not attainable, and some forms of knowledge defy
> capture, KR should focus on reconciling insights from mathematics,
> philosophy, theoretical physics, cognitive and neuroscience,
> psycholinguistics to name a few, keeping in mind that any formalized
> systems. whether logical, natural language based, mathematical or
> computational are inherently incomplete.
>
> Milton Ponson
> GSM: +297 747 8280
> PO Box 1154, Oranjestad
> Aruba, Dutch Caribbean
> Project Paradigm: Bringing the ICT tools for sustainable development to
> all stakeholders worldwide through collaborative research on applied
> mathematics, advanced modeling, software and standards development
>
>
> On Monday, August 19, 2019, 11:40:22 PM EDT, Paola Di Maio <
> paola.dimaio@gmail.com> wrote:
>
>
> Sharing Philip's offlist reply with valuable website and publications
>
> It really inspires to publish a personal website, should do the same
>
> I find a lot of the AGI I hear about is sill very narrow in a general sense
>
> Maybe Philip's can help us to move on away from narrow AGI
> A lot to read!!
>
> PDM
>
>
>
> ---------- Forwarded message ---------
> From: *Philip Jackson* <philipcjacksonjr@hotmail.com>
> Date: Mon, Aug 19, 2019 at 9:23 PM
> Subject: NL, formal logic, machine language, etc.
> To: paola.dimaio@gmail.com <paola.dimaio@gmail.com>
>
>
> Hello Paola,
>
> This topic is discussed somewhat in my thesis and subsequent papers,
> available via the Publications page at www.talamind.com.
>
> Best,
>
> Phil Jackson
>
>