Newsgroups: comp.ai.genetic
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!howland.reston.ans.net!cs.utexas.edu!utnut!alchemy.chem.utoronto.ca!jdods
From: jdods@alchemy.chem.utoronto.ca (Jeffrey Dods)
Subject: Re: Is continuous variable optimization  really a problem in GAs?
Message-ID: <1995Jan20.190256.942@alchemy.chem.utoronto.ca>
Date: Fri, 20 Jan 1995 19:02:56 GMT
References: <18JAN199509144578@sigma.tamu.edu>
Organization: University of Toronto Chemistry Department
Lines: 20

In article <18JAN199509144578@sigma.tamu.edu>,
HANAGANDI, VIJAYKUMAR M. <vmh0648@sigma.tamu.edu> wrote:
[...]
>a population, the mapping from these strings to the real numbers
>might yield inaccuracies depending on how long is the binary 
>string (the longer it is the better is the accuracy).
>this seems to be of concern when the parameters to be optimized
>are reals.
>are there any modifications/tricks to circumvent this problem?
>
>any suggestions, comments are appreciated.
>
>-Vijay
>Los Alamos.

Have a look at Schraudolph's modification of GENESIS called GAucsd. 
The ftp site for it seems to keep changing, try mailing nici@cs.ucsd.edu.
 -good luck.
 Jeff.

