🚀 go-pugleaf

choup@ava.bcc.orst.edu (Ping Chou) writes:
>
>        Or the development of genetic algorithms has completed?
>
>        Isn't it true that selection/mating/mutation/chromosome/hybrid/
>genotype/phenotype/gene/... are taken from genetics?
>
>        Don't you think you may be able to discover new algorithms from
>studying genetics?
>
>        Do you really believe the complete genome of HIV is just a 4-state
>bits encoded string?
>
>        Then please, give youself a chance, take a look at any modern
>genetics text book, and be a cross-discipline expert.
>
>--Ping
>
>        BTW, I would appreciate any reasonable arguments because I'm trying
>to learn ai-genetic from the genetics side.


Dear Ping and anyone else who agrees with him,

I would like to agree with some of the other responses I have read and
call you arrogant, but I suspect that you probably get some strange
sort of pleasure from this so I won't (except that I already did -
Oh well). 

Anyway what I wanted to say was that I think you are being misled by
the name "Genetic Algorithms" which I consider not to be a very good
one (am I being arrogant now?).  Genetics is only one place where
evolution is visible.  It happens to be one which is perhaps easier
to analyse in a scientific way than others, but nevertheless if you
restrict your thoughts to this field only you are shooting yourself
in the foot.

The genetic model for what could be called "Artificial Evolution"
imposes unnecessary restrictions on development.  If you were to argue 
that it doesn't you'd be arguing that no operation exists which could be 
useful yet is not genetically viable.  At the same time you'd also be
contradicting yourself since evolution is all about trying new things
and letting variety prosper, and hence you'd be restricting your creativity
because of some gospel-like belief that genetics is the true model for
artificial evolution.

Can I suggest that you think of the chromosome (as termed in GAs), not as 
a chromosome but as a representation of an entity, and that this entity
can be anything which can evolve, be it an economic theory, a political
system or a biological organism.

I am not discounting the usefulness of the genetic model as a learning
tool but I am disputing that it be regarded as the "God of Evolution".


Hans Andersen
University of Queensland
Australia