From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 14987 invoked by alias); 23 Jan 2008 07:26:26 -0000 Received: (qmail 14970 invoked by uid 22791); 23 Jan 2008 07:26:25 -0000 X-Spam-Check-By: sourceware.org Received: from mailsend02.psi.ch (HELO MAILSEND02.psi.ch) (129.129.191.54) by sourceware.org (qpsmtpd/0.31) with ESMTP; Wed, 23 Jan 2008 07:25:52 +0000 Received: from MAILBOX0A.psi.ch ([129.129.190.193]) by MAILSEND02.psi.ch with Microsoft SMTPSVC(6.0.3790.3959); Wed, 23 Jan 2008 08:25:49 +0100 Received: from 129.129.206.141 ([129.129.206.141]) by MAILBOX0A.psi.ch ([129.129.190.181]) via Exchange Front-End Server mail.psi.ch ([129.129.191.54]) with Microsoft Exchange Server HTTP-DAV ; Wed, 23 Jan 2008 07:25:48 +0000 MIME-Version: 1.0 Received: from pc6411 by mail.psi.ch; 23 Jan 2008 08:25:48 +0100 Content-Type: multipart/signed; boundary="=-KnRvzMqpC/N/VLrpgs6D"; micalg=pgp-sha1; protocol="application/pgp-signature" Content-class: urn:content-classes:message Subject: Re: multidimenstional minimization without df. Date: Wed, 23 Jan 2008 07:26:00 -0000 Message-ID: <1201073148.19829.13.camel@pc6411.psi.ch> In-Reply-To: <4795F855.6010105@physics.gatech.edu> References: <4795F855.6010105@physics.gatech.edu> From: "Alxneit-Kamber Ivo" To: "Toan T Nguyen" Cc: Reply-To: "Alxneit-Kamber Ivo" Mailing-List: contact gsl-discuss-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: gsl-discuss-owner@sourceware.org X-SW-Source: 2008-q1/txt/msg00020.txt.bz2 --=-KnRvzMqpC/N/VLrpgs6D Content-Type: text/plain Content-Transfer-Encoding: quoted-printable Content-length: 1750 On Tue, 2008-01-22 at 09:06 -0500, Toan T Nguyen wrote: > Hi, >=20 > I'd like to locate the minimum point of a function f(x_i) of n variables= =20 > x_i. It's very non trivial to calculate the df/dx_i. Can I use conjugate= =20 > gradian method without this df(x_i) information ? >=20 > This page in the manual=20 > http://www.gnu.org/software/gsl/manual/html_node/Multimin-Algorithms.html >=20 > doesn't list gsl_multimin_fminimizer_conjugate_fr as a type of=20 > minimizer. It has > gsl_multimin_fdfminimizer_conjugate_fr > but no > gsl_multimin_fminimizer_conjugate_fr >=20 > Any help would be appreciated. >=20 > Thanks, > Toan hi, no, the conjugate gradient method needs the derivatives (to calculate the gradient). thus, there can be no minimizer gsl_multimin_fminimizer_conjugate_fr (fdf minimizer: you supply f(x) AND df/dx. f minimuzer: you only supply f(x)). so, either you are able to supply the derivatives or you will end up with gsl_multimin_fminimizer_nmsimplex that needs no derivatives. i often use gsl_multimin_fminimizer_nmsimplex to fit parameters of rather complex simulations. here the function evaluation takes up to a few minutes and i do not have access to the derivatives. note, that i have applied following patch http://www.network-theory.co.uk/download/gslextras/SimplexImproved/ (otherwise the minimizer most often gets stuck at the starting point) and i scale the parameters in a way to make their numerical values similar (i usually divide by their initial values). --=20 Dr. Ivo Alxneit Laboratory for Solar Technology phone: +41 56 310 4092 Paul Scherrer Institute fax: +41 56 310 2688 CH-5232 Villigen http://solar.web.psi.ch Switzerland gnupg key: 0x515E30C7 --=-KnRvzMqpC/N/VLrpgs6D Content-Type: application/pgp-signature; name=signature.asc Content-Description: This is a digitally signed message part Content-length: 189 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (GNU/Linux) iD8DBQBHluv6Ad7CE1FeMMcRAjcuAKClhk3mA+ZyHdHUYV+xl6RbWGgg3wCcDN8v /FPuw1xNMBV7CJbu+PbEujM= =6+Ii -----END PGP SIGNATURE----- --=-KnRvzMqpC/N/VLrpgs6D--