--- Jim Bromer <[EMAIL PROTECTED]> wrote:
> Let's say that a method (might as well say my method) produces a solution to
> the SAT problem where even the worse cases can be determined in polynomial
> time.
> 
> Now let's say that the problem is not to determine whether or not there is a
> solution or not, but to output every possible combination of values for a
> logical formula and give the logical result as well.  Since there are 2^n
> combinations for a formula of n distinct logical variables, the output would
> have to take more than 2^n 'steps'.  So even supposing the np-complete
> problem could be solved in polynomial time, there are still problems that
> cannot be completed in polynomial time.

This is an example of a problem which is not in NP.  A problem is in NP if a
solution can be verified in polynomial time.  A problem is in P if the
solution can be found in polynomial time.  By polynomial, I mean that in the
worst case, the time is a polynomial function of the size of the input.  For
example the SAT problem is to determine if there is a variable assignment that
evaluates to true, and if so, output a variable assignment.  The traveling
salesman problem (TSP) is, given a graph and a number D, to determine if there
is a cycle of length D or less that visits each vertex exactly once, and if
so, output the cycle.

SAT and TSP are NP-complete.  That means any problem in NP can be solved by
calling one of these problems as a subroutine by a program that is otherwise
in P.  Thus, if SAT or TSP is in P, then all problems in NP are also in P, in
which case P=NP.

There are over 3000 problems known to be NP-complete.  If any one of them can
be solved in polynomial time, then they all can.  It is widely believed that
P!=NP because a lot of people have tried and failed to do this.  However, it
is not proven that P!=NP.  The Clay Institute has offered a $1 million prize
for a proof either way.  A partial list of problems can be found here:
http://en.wikipedia.org/wiki/List_of_NP-complete_problems

Good luck.


-- Matt Mahoney, [EMAIL PROTECTED]

-----
This list is sponsored by AGIRI: http://www.agiri.org/email
To unsubscribe or change your options, please go to:
http://v2.listbox.com/member/?member_id=8660244&id_secret=87672524-64fc61

Reply via email to