> well at least with love there is reproduction.
> SAT seems fairly limited domain.
>
> That's why I rather work towards robots that can replicate.
> Logan


I am not totally sure how SAT in p would be worked into AGI even
though I am very interested in both of the problems. I assume that as
new computational theories are developed in response to poly time SAT
we would be able to use them in AGI theories. But if I discovered a
way to solve SAT in polynomial time I would not have a clear idea
about how I could use that in an AGI project. However, I think I do
have a clue. In my opinion, variations of the frame problem are
central to the problem of contemporaneously testing out different AGI
models. And it has always been my understanding that the frame problem
is something that SAT in p would help with. My best personal
explanation for this intuition is that I could use multiple 'levels'
(virtual and in many cases temporal levels) of relations in my AGI
project. These 'levels' would, in many cases, simplify the problems of
complex relations. Weighted Reasoning is another way that complex
relations can be used in -some- situations to simplify complex
relations. But what I am trying to say is that multiple level
interrelations are as important as weighted reasoning but it is no
more a complete solution to the problems posed by the frame problem
than Weighted Reasoning is. Now if there was a polynomial time
solution to SAT then additional cross categorical methods (more
'levels' that connect across other 'levels') could be developed and
more quickly tested for logical consistency. The goal is not to
achieve computational perfection but to get beyond the current
low-level of SAT to find out how the theories about the world (that
the AGI program would create) might be connected and interrelated.
Jim Bromer


On Thu, Feb 12, 2015 at 7:05 AM, Logan Streondj <[email protected]> wrote:
> On Wed, Feb 11, 2015 at 03:35:49PM -0500, Jim Bromer via AGI wrote:
>> There is a Turkish proverb that says, "Before you love, learn to run
>> through snow without leaving footprints."
>>
>> So finding a deterministic polynomial time solution to SAT is, at
>> least in some ways, like love.
>>
>> Jim Bromer
>
> well at least with love there is reproduction.
> SAT seems fairly limited domain.
>
> That's why I rather work towards robots that can replicate.
>
> --
> Logan


-------------------------------------------
AGI
Archives: https://www.listbox.com/member/archive/303/=now
RSS Feed: https://www.listbox.com/member/archive/rss/303/21088071-f452e424
Modify Your Subscription: 
https://www.listbox.com/member/?member_id=21088071&id_secret=21088071-58d57657
Powered by Listbox: http://www.listbox.com

Reply via email to