* Optimization in GSL
@ 2002-10-30 13:05 Lukasz Grabun
2002-10-30 13:17 ` Lukasz Grabun
0 siblings, 1 reply; 2+ messages in thread
From: Lukasz Grabun @ 2002-10-30 13:05 UTC (permalink / raw)
To: gsl-discuss
OK, that's a big one. Is there a support for quadratic optimization in
GSL? I've browsed the documentation and then read it rather carefully
but I just could not find any routines that would solve something like
f(x) = \sum_{s,i} (x_{s,i}^2 - a_{s,i})^2 subject to constraints
x_{s,i} >= 0;
\sum_s x_{s,i} = 1;
It's very simple problem thought due to number of variables one can
not solve this manually. I though I can use computer for this purpose.
Did anyone happen to know how one can use GSL to solve such a problem?
--
Lukasz Grabun
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: Optimization in GSL
2002-10-30 13:05 Optimization in GSL Lukasz Grabun
@ 2002-10-30 13:17 ` Lukasz Grabun
0 siblings, 0 replies; 2+ messages in thread
From: Lukasz Grabun @ 2002-10-30 13:17 UTC (permalink / raw)
To: gsl-discuss
On Thu, Oct 31, 2002 at 08:50:15PM +0100, Lukasz Grabun wrote:
> f(x) = \sum_{s,i} (x_{s,i}^2 - a_{s,i})^2 subject to constraints
Ouch, my wrong.
f(x) = \sum_p [ \sum_{s,i} x_{s,i} * a_{s,p,i} ]^2
a_{} = a constant number, generally a_{} = +/- 1 (yes, it does
originate from Ising problem).
Constraints does not change. Sorry for misleading anyone.
--
Lukasz Grabun
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2002-10-30 21:15 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2002-10-30 13:05 Optimization in GSL Lukasz Grabun
2002-10-30 13:17 ` Lukasz Grabun
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).