I am just about giving up on my attempt to create a deterministic
polynomial time solver for SAT. I am still working on one conjecture,
even though it is in np, because it is interesting. It uses the
artifact of a truth table to enumerate the rows of unique states of
the literals (the logical variables). Solutions (and non-solutions)
then can be represented by numerical patterns. I am working on 3-SAT.
The method is very difficult for me because I haven't found the
fundamental algebra to work the representation of the patterns, so I
am constantly generating some of the patterns in order to understand
how to work the next step of the solution out.

So I have found no rational evidence that I received any divine
inspiration to work on this problem, which was a possibility that I
considered in 2007.

Although the system will still be in np even when I work the algebra
out, there is still a slight possibility that I might move on to the
next step and create an algebra for what I call the fractionation of
the generative patterns. As the generative patterns intersect the
extent of the intersections becomes fractionated. If I can discover an
algebra for that effect then I might be able to discover a pattern
generator for the fractionation.  So I believe that the problem might
be approachable by applying (at least) two levels of pattern
generators - but only if I can derive an algebra for the different
levels.

However, I am not hoping to derive a solution that is in p. Although I
clearly was motivated by that goal, I am more interested in working on
more feasible aspects of the problem now. There has been some progress
made on the problem in the last decades and maybe I will one day be
able to add something to that history.
Jim Bromer


-------------------------------------------
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