[R] optimization free software

Prof. John C Nash nashjc at uottawa.ca
Fri Aug 21 16:25:30 CEST 2009


There are also several projects on r-forge.r-project.org that deal with 
optimization, including one I'm involved with that has been putting up a 
number of new or reworked methods so that they can be tested, evaluated 
and improved. These all deal with unconstrained or box-constrained 
minimization problems (minimize -f(x) to maximize). In particular, we 
have an implementation of Mike Powell's BOBYQA (derivative-free 
box-constrained method using quadratic approximation) that seems quite 
interesting. See package minqa in the project 
http://r-forge.r-project.org/R/?group_id=395.

There is also a reworking of conjugate gradients with a rather better 
automated choice of generation formula (using Dai and Yuan, 2001 ideas) 
with box-constraints. It handles very large problems, often amazingly 
quickly e.g., an n=5000 generalized Rosenbrock problem, with analytic 
gradients, in <3 secs on an Asus Eee 900 netbook. And that is an all-R 
implementation.

Caveat: all these are developmental codes. We believe they "work", but 
we are trying to torture test them and are finding some situations where 
they have glitches. We welcome use and comments to improve them. I 
expect the same applies to other codes on r-forge.

John Nash




Message: 98
Date: Thu, 20 Aug 2009 11:08:29 -0700
From: Gaspar N??ez <nurogaleo at gmail.com>
Subject: [R] optimization free software
To: r-help at r-project.org
Message-ID:
	<9cd4bb290908201108x554ad8a3h520417208a3304b3 at mail.gmail.com>
Content-Type: text/plain

Hi

i´m starting to work with free software
(with free as in freedom)

i've been working with GAMS
(General Algebraic Modeling System)
to solve static and dynamic optimization problems
and non-linear large systems of equations

i'm wondering if there is any free software
and documentation on the subject,
wether in R, Maxima, or something else

i will appreciate any help
and any references about this issue

thanks in advance

gaspar




More information about the R-help mailing list