Ben et al

For what it's worth, you might want to reconsider my rudimentary work on tacit 
knowledge engineering, which deals extensively with association and lower-level 
functions and its many complications within architectural constraints (as the 
CAMF). (freely available on researchgate).

For excellent reasons, I never published any white papers on the method after 
2008. Still, R&D never stopped. The "missing" work all the way to genetic 
engineering has been completed. Further, to my mind, the world of "association" 
and all its complexities has also been conquered, even down to solving the 
most-relevant riddle of 'priority of priorities'.

Based on a useful ontology of natural language, which I see graph theory relies 
on most similarly, and associative functionality, the struggle to extend the 
method into operational functionality and how to derive the knowledge contexts 
from reverse-engineered functionality, was real. I think I spent a year on that 
aspect before "discovering" a working solution for the method specifically (not 
for all of science). Still, the work is never done.

I'm definitely no mathematician, but if I understood correctly what you're busy 
talking about, the latest extention to the method - still in theoretical 
development, as The Pattern of One (Po1) - has opened a doorway to potentially 
formalizing Riemann functionality, perhaps even to autonomous quantum 
encryption and messaging, etc.

I have my genius, and everyone else has his/her own. It's not a competition, or 
at least, it should not be. Diversity is a key in itself. Perhaps, it's time to 
pool diverse genius, with due regard for IP?

Point remains, the quantum-based method still hasn't failed after all these 
years of field testing, and if the hypothesis of seamlessly-scaling concurrent 
engineering in near-real time proves correct, it shouldn't either. My wild 
guess is that when quantum computing becomes available, this would be the key 
constraint developers are going to encounter in their logic. It doesn't have to 
be this way.

My recent "From NGI (Natural General Intelligence) to AGI; troubled 
thoughts..." article on LinkedIn should give all AGI researchers and developers 
pause to consider. I was going to publish it on this list, but I decided a 
broader audence would be more appropriate.

Regards

Robert Benjamin
________________________________
From: Ben Goertzel <[email protected]>
Sent: Tuesday, 09 February 2021 02:39
To: AGI <[email protected]>
Subject: Re: [agi] new paper: Logic in Hilbert space


I wrestled with similar issues regarding associativity and cognitive operations 
when writing this paper

https://arxiv.org/abs/2004.05269

-- see Section 6.2.1 where I show that if you have a set of mutually 
associative combinational operators, then you get a very nice subpattern 
hierarchy ... which is much messier to get without mutual associativity...

An interesting question is whether: IF you have a set of mutually associative 
combinational operators, can you then isomorphically map the functions being 
combined into Hilbert space vectors to that the (associative) combinations map 
into vector operations?   I have not thought about this before and am not sure 
if it can be made to work

Quantum mechanics uses complex Hilbert spaces, and i did go thru some work to 
make an approximate isomorphism btw paraconsistent logic truth values and 
complex amplitudes,

https://arxiv.org/abs/2101.07498

but that addresses a different part of the same problem, it seems

-- Ben




On Sun, Jan 31, 2021 at 2:45 AM YKY (Yan King Yin, η”„ζ™―θ΄€) 
<[email protected]<mailto:[email protected]>> wrote:
On 1/30/21, Ben Goertzel <[email protected]<mailto:[email protected]>> wrote:
> Unless I remember wrong (which is possible), function application in a
> Scott domain is not associative, e.g.
>
> (f(g) ) (h)
>
> is not in general equal to
>
> f( g(h) )
>
> However function composition is associative, and the standard products
> on vectors in Hilbert space are associative
>
> So it seems what you're doing may not be quite right, and you need to
> be looking at some sort of fairly general nonassociative algebras over
> Hilbert space instead ... or something...
>
> Or am I misunderstanding something?


Damn... you're right πŸ˜†

I have no idea how to make one function "apply" to another, except by
function composition, so that was what I did.  But I forgot about
associativity...

I don't know if I should pursue along this any further.... it seems
computer-implementable but it's very complicated...  and I currently
don't know how to make non-associative algebras....

Thanks for pointing out my mistake, it's helpful πŸ˜…
YKY


--
Ben Goertzel, PhD
http://goertzel.org

β€œHe not busy being born is busy dying" -- Bob Dylan
Artificial General Intelligence List<https://agi.topicbox.com/latest> / AGI / 
see discussions<https://agi.topicbox.com/groups/agi> + 
participants<https://agi.topicbox.com/groups/agi/members> + delivery 
options<https://agi.topicbox.com/groups/agi/subscription> 
Permalink<https://agi.topicbox.com/groups/agi/T54594b98b5b98f83-M5d4d8e1d1e09d9b49c528cc7>

------------------------------------------
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/T54594b98b5b98f83-Mec971db895f3ba814a9f091d
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Reply via email to