Newsgroups: comp.ai.alife
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!news.mathworks.com!hookup!swrinde!pipex!sunic!news.chalmers.se!news.gu.se!gd-news!d6250.shv.hb.se!sa209
From: sa209@utb.shv.hb.se (Claes Andersson)
Subject: Re: Evolvable Fitness Formula
Message-ID: <sa209.87@utb.shv.hb.se>
Sender: usenet@gdunix.gd.chalmers.se (USENET News System)
Nntp-Posting-Host: d6250.shv.hb.se
Organization: Department of Scocial Science
References: <3bt1am$q25@srvr1.engin.umich.edu>
Date: Thu, 8 Dec 1994 17:16:55 GMT
Lines: 51

In article <3bt1am$q25@srvr1.engin.umich.edu> streak@engin.umich.edu (Theodore C. Belding) writes:

>In article <sa209.81@utb.shv.hb.se> Claes Andersson, sa209@utb.shv.hb.se 
>writes:
>> It think it is important to make a distinction between simulation of life 
>>and utilizing lifelike systems for optimizations. The entire landscape model 
>>falls when the fitness formula is emergent from the animates traits. This is 
>>because all animates is located at one place each in a landscape that is 
>>defined by where it is itself and where all others is. The landscape 
>>determines the landscape.. the hen and the egg.. or a paradox. Well, not 
>>quite but still meaningless.

>You may be right that there are other models than the fitness landscape
>that are better in certain situations, but it is hardly meaningless, or
>anything like a paradox.  When he came up with the idea, Sewall Wright
>was perfectly aware that the landscape was dependent upon interactions
>between the organisms, each other, and the environment.

>Picture a rubber sheet.    This is the fitness landscape.  
>If you like you can put deformations in the sheet to produce fitness
>ridges, valleys, etc.  Now place some weighted balls on the sheet.  
>These are the effects on the landscape caused by different groups of 
>organisms.  As the balls move on the sheet (the landscape), they deform
>the sheet, causing the balls to move in other directions, causing more
>deformations...

>This is the standard model of the deformation of time and space caused
>by gravity.  It is perfectly applicable to Sewall Wright's fitness landscapes
>(sure, it's simplified -- the effects of organisms on the landscape
>don't act like balls, but the general idea is valid).  It is simply
>a dynamic system with high degrees of non-linearity.
>-Ted

 As you say, the general idea is valid and very useful. It's absolutely not 
meaningless in all situations, far from. What I ment was that the increase 
in complexity when the fitness formula is derived from traits is tremendous. 
A simple example could be, I imagine, to let them be eat eachother to get 
nutrition. What will happen is that, quouting an article in Scientific 
American that I don't remember who wrote, the lion's genes for long, sharp, 
teeth affects the length and strength of the zebras legs indirectly. There 
is hardly no distinction between the genomes of different specimens, the 
only distinction is between the abstract genes.

 The landscape model is in this context still useful but impossible to study 
over time since it will only be valid in exactly one point on the time-axis 
at a time. It builds on what it describes. When the fitness formula is fixed 
and you have two genes, it is totally valid but it is hardly valid in a 
natural context. Correct me if I'm wrong.


Claes Andersson. University of Bors. Sweden
