Re: Algorithm for solving system of equations?
- From: Ray Vickson <RGVickson@xxxxxxx>
- Date: Mon, 28 Jul 2008 07:53:39 -0700 (PDT)
On Jul 28, 3:04 am, perkins.har...@xxxxxxxxx wrote:
I know that solving a system of equation is "difficult" (in NP).
However, I have a dynamic system consisting of 56 equations with 56
unknowns, and am trying to come up with a good algorithm for solving
these equations. I can technically have 64 equations with 56
unknowns, but that doesn't really seem to make my life any easier on
the reduction front.
Does anyone know a good algorithmic way of doing this?
Technically, if the equations are *nonlinear* the problem is not even
in NP. (For example, there is no finitely-terminating algorithm to
solve the equation x^2 = 2 numerically.) However,the main point is
that the problem is hard. Is this for an application in which you
know, somehow, that the equations DO have a solution? I would suggest
that you present a few more details here, since issues like
"structure", "sparseness", etc., can make a tremendous difference to
how you should proceed.
R.G. Vickson
.
- References:
- Algorithm for solving system of equations?
- From: perkins . harold
- Algorithm for solving system of equations?
- Prev by Date: Re: help solving a textbook variational problem
- Next by Date: Re: an important set theory post
- Previous by thread: Re: Algorithm for solving system of equations?
- Next by thread: Finite fields
- Index(es):
Relevant Pages
|