Newsgroups: comp.ai.genetic
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!howland.reston.ans.net!math.ohio-state.edu!jussieu.fr!univ-lyon1.fr!swidir.switch.ch!newsfeed.ACO.net!edvz.sbg.ac.at!news
From: helmut@cosy.sbg.ac.at (Helmut Mayer)
Subject: Re: Simple question..
Message-ID: <D8tuqn.LEw@wst.edvz.sbg.ac.at>
Sender: news@wst.edvz.sbg.ac.at (USENET News System)
Reply-To: helmut@cosy.sbg.ac.at
Organization: University of Salzburg / Austria
References: <D8s0sp.9Hx@curia.ucc.ie>
Date: Fri, 19 May 1995 13:27:59 GMT
Lines: 42

Mike McElligott writes

# 	Is it correct to say that rank based selection is a form of fitness
# scaling ?

Depends..;-) If you are concerned with the goals, namely, avoiding premature  
convergence and slow finishing, of employing these techniques, they are  
similar. The technique itself differs in that with fitness scaling you map a  
"raw fitness" to a scaled fitness but still operate with fitness values from  
which the selection probability is derived. Whereas, with rank based methods  
you map the "raw fitness" to a rank within the generation and from there on  
don't care about the fitness values, i. e. the information about "fitness  
distance" is lost.
You could take a look at:

@INPROCEEDINGS(Baker85,
	AUTHOR = "James Edward Baker",
	TITLE = "Adaptive Selection Methods for Genetic Algorithms",
	BOOKTITLE = "Proceedings of the First International Conference on  
Genetic Algorithms and their Applications",
	YEAR = 1985,
	EDITOR = "John J. Grefenstette",
	PAGES = "101--111",
	ORGANIZATION = "Texas Instruments, Inc. and Naval Research Laboratory",
	PUBLISHER = "Lawrence Erlbaum Associates")

@INPROCEEDINGS(Maza93,
	AUTHOR = "Michael de la Maza and Bruce Tidor",
	TITLE = "An Analysis of Selection Procedures with Particular Attention  
Paid to Proportional and Boltzmann Selection",
	BOOKTITLE = "Proceedings of the Fifth International Conference on  
Genetic Algorithms",
	YEAR = 1993,
	EDITOR = "Stephanie Forrest, University of New Mexico",
	PAGES = "124--131",
	ORGANIZATION = "University of Illinois at Urbana-Champaign",
	PUBLISHER = "Morgan Kaufmann")

Best regards,

Helmut

