MindForth First Working AGI needs logical premises.

[menti...@gmail.com]

The free-of-charge, open-source MindForth program is the First Working Artificial General Intelligence (AGI) because it has been thinking with concepts (not statistics) for the past ten years since 2008. It solves the hard problem of http://en.wikipedia.org/wiki/Natural_language_understanding -- NLU -- and it engages in automated reasoning with logical inference. If MindForth knows that certain entities do certain things, such as "STUDENTS READ BOOKS", you may type in "john is a student" and the First Working AGI will ask you, "DOES JOHN READ BOOKS".

http://ai.neocities.org/MindBoot.html -- is the MindBoot sequence which contains enough innate knowledge and English vocabulary to get the First Working AGI up and running when you initiate the program. MindBoot contains one logical premise in the idea of "WOMEN HAVE A CHILD" but it needs more such starting points for logical inference. We could add "MEN HAVE A WIFE" so that if a user enters "jim is a man", the First Working AGI could ask, "DOES JIM HAVE A WIFE". Students and AI Mind Maintainers need more such logical premises either for insertion into the MindBoot or for teaching to the AI as part of demonstrating logical inference. Already some candidate premises are:

- Persons own pets.
- Doctors play golf.
- Farmers drive trucks.

Cheers,

Mentifex (Arthur)
--
http://ai.neocities.org/mindforth.txt -- source code in Win32Forth;
http://cyborg.blogspot.com/search/label/MindForth -- programming journal;
http://dl.acm.org/citation.cfm?doid=307824.307853 -- MindFO\orth ACM paper;
http://www.amazon.com/dp/B00FKJY1WY -- InFerence Kindle e-book.

[minf...@arcor.de]

Am Montag, 2. Juli 2018 12:56:23 UTC+2 schrieb menti...@gmail.com:
> The free-of-charge, open-source MindForth program is the First Working Artificial General Intelligence (AGI) because it has been thinking with concepts (not statistics) for the past ten years since 2008. It solves the hard problem of http://en.wikipedia.org/wiki/Natural_language_understanding -- NLU -- and it engages in automated reasoning with logical inference. If MindForth knows that certain entities do certain things, such as "STUDENTS READ BOOKS", you may type in "john is a student" and the First Working AGI will ask you, "DOES JOHN READ BOOKS".
>
> http://ai.neocities.org/MindBoot.html -- is the MindBoot sequence which contains enough innate knowledge and English vocabulary to get the First Working AGI up and running when you initiate the program. MindBoot contains one logical premise in the idea of "WOMEN HAVE A CHILD" but it needs more such starting points for logical inference. We could add "MEN HAVE A WIFE" so that if a user enters "jim is a man", the First Working AGI could ask, "DOES JIM HAVE A WIFE". Students and AI Mind Maintainers need more such logical premises either for insertion into the MindBoot or for teaching to the AI as part of demonstrating logical inference. Already some candidate premises are:
>
> - Persons own pets.
> - Doctors play golf.
> - Farmers drive trucks.
>

It seems like a reinvention of Prolog.

[Liang Ng]

On Monday, 2 July 2018 21:25:21 UTC+8, minf...@arcor.de wrote:
> It seems like a reinvention of Prolog.

Prolog cannot do OpenGL.

Forth and variants can.

[gnuarm.del...@gmail.com]

Are you suggesting logical premises? Interesting suggestions.

Rick C.

[minf...@arcor.de]

Am Dienstag, 3. Juli 2018 14:39:06 UTC+2 schrieb Liang Ng:
> On Monday, 2 July 2018 21:25:21 UTC+8, minf...@arcor.de wrote:
> > It seems like a reinvention of Prolog.
>
> Prolog cannot do OpenGL.
>

My.. another nonsense from you. Google yourself.