My latest theory has greater representative power than traditional 3-SAT
logical methods and it was exciting to see that. However, I haven't been
able to get a good idea about the classes of logical relations that it does
represent well so I cannot yet tell if it can actually handle 3-SAT yet. I
mean I know I can make it work in NP but I can't tell if there is a
significant improvement even though there is a significant improvement
(over representative range) for simple logical problems.

For example, there are simple ways that you can add new logical relations
to cover any logical statement with three variable literals in it. Once you
do that logical transformations can be made to any logical statement that
uses three variable literals no matter how complicated so that you can
reduce the problem to a single logical function over those three variable
literals. However, at each higher level logical problem, where there are
more than three variable literals, there are logical relations which cannot
be reduced to a 1, 2 or 3 variable logical function. But
the improvement would extend to certain classes of problems in 4, 5, 6,
..., n logical variables, because there are some logical formulas which can
be expressed by using equivalences that are expressible in 3 variables.

So I can't tell if my system is like extending the logical functions to
include all possible relations that can be expressed in 3 variables (for
example) or if the improvements in representative range are more like
functions that can be extended for other kinds of classes of problems. If
my system was able to show actual gains for classes of problems that were
different than can be found using simple extensions for 3, 4, 5, (or some
n) problems then it might be a true innovation that should be examined
carefully for further development.



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