Newsgroups: comp.ai.genetic
Path: cantaloupe.srv.cs.cmu.edu!bb3.andrew.cmu.edu!newsfeed.pitt.edu!scramble.lm.com!news.math.psu.edu!news.cse.psu.edu!uwm.edu!math.ohio-state.edu!howland.reston.ans.net!newsfeed.internetmci.com!in2.uu.net!sangam!iitb!powai!skypak.aero.iitb.ernet.in!pbn
From: pbn@I_should_put_my_domain_in_etc_NNTP_INEWS_DOMAIN (Prasanth Nair)
Subject: Re: GA and large search spaces
Sender: news@powai.cc.iitb.ernet.in (news system user)
Message-ID: <DtrAFF.4K4@powai.cc.iitb.ernet.in>
Date: Sat, 29 Jun 1996 09:59:38 GMT
References: <Dto3GL.F9K@exeter.ac.uk>
Organization: Indian Institute of Technology, Bombay
X-Newsreader: TIN [version 1.2 PL2]
Lines: 28

Dragan Savic (D.Savic@exeter.ac.uk) wrote:
: I am working on a problem which, if represented using binary coding, will 
: result in strings of 500 up to 1000 bits. One possible option is to use 
: Messy GAs. Does anyone out there has any useful ideas or knows about 
: similar problems solved by using GAs or hybrid techniques. Tips, 
: hints or anything of help will be appreciated. Thanks in advance.

: Regards,
: Dragan Savic

For such string sizes, the convergence of binary coded GAs may be extremely 
slow. Two possible alternatives I can think of (assuming your variables 
are continuous) :

(1) Use a real coded GA

(2) GAs with gradient life cycles => A Genetic Gradient Hybrid
    
    Reference : Flasse and..., 1996, "Hybrid Genetic Gradient Algorithms 
    for Global Optimization".
    IEEE Transactions : Systems, Man and Cybernetics. 
   		
   I don't recollect the exact vol/issue..most probably in April/ May
   
 Regards 	
 P.B.Nair
 	
   	
