--- Dave Griffiths <dave_at_pawfal.org> wrote:
> On Thu, 31 Mar 2005 01:46:34 +0200, Julian Rohrhuber
> wrote
> > >Thinking about it some more, and discussing this
> off list - I think that
> > >this is actually a pretty promising idea.
> > >
> > >You could livecode a fitness function that would
> select from populations
> > >of genetic programs that create music.
> > >
> > >Makes my head hurt tho :)
> > >
> > >dave
> >
> > I think this is an interesting topic to discuss
> on-list. Aren't
> > languages ways to make things thinkable that
> weren't otherwise?
ok. on-list!
As I was talking to dave, I'm thinking about the
subject of my master thesis(I'm just begining so I
very very excited about it). At the University of São
Paulo.
And then I had the ideia of putting genetic
programming and live programmed music together. I sent
a message to this list and dave replied me...
>
>... cross-over ...
usualy in GP the programms are displayed in their
parse-tree form. A cross-over happens when branches
are axchanged between direrent (or not) parse trees.
>
> In a simple system like my GP synth, the selection
> is manual - I'm stepping
> though the "semi infinite space of possible noises"
> manually, and according to
> my own whimsy. The populations are only ever 4 big.
There is "size" problem yep... How could I show
several parse trees (of the running programms) in a
friendly user(audience and programmer) interface?
I thought about a tree of trees or a list of trees...
???
>
>
> ...a fitness function ...
There is this() paper where the guy talks about
something like what we are thinking of but not live.
The user has to wait until one music ends, rate it,
and then move to another.
http://graphics.stanford.edu/~bjohanso/gp-music/
He realized that this was slow and boring and tried to
use a trained neural network to do the fitness
function. I'm not sure that this is nice becouse it's
like shooting your own foot.. Using a meta-heuristic
to give data to another...
I thought initialy an easy way to construct the
fitnnes function would be:
Each programm would have a probability of survival. At
the end of a turn a random mechanism would decide if a
programm would survive to reproduce or die according
to the probability..
If the programmer does not set a probability of
survival to a programm then we can do lots of things
like it inherits an average of its parents
probabilities...
Usualy the fitness function is refered as an important
part of the "enviroment" where the indiviuals are
living. We could say that in a "live coding gp system"
(lcgps(hehehe)) the enviroment would be the
music-programmer! He can edit a given proggram, he can
kill anyone that's bothering....
>
> ramble ramble?
>
marcel
Send instant messages to your online friends
http://uk.messenger.yahoo.com
Received on Fri Apr 01 2005 - 03:19:06 BST