heads up

[Gerald de Jong]

I've been talking to Joost Rekveld (http://www.lumen.nu/rekveld/wp/)
about a date for the second Greythumb NL gathering, and we're thinking
about the last part of september, at his academie in Den Haag.

I'd like to ask you to think about what you might present or talk
about, if you've ever thought about standing up. The bar isn't very
high. Even stories about a few evenings spent playing with an
existing artificial life simulation would be great! We can all learn
from that. This is supposed to be fun.

I was hoping to hear from Anne van Rossum about his project, focusing
on what is biologically inspired or derived.

From a very interesting biota conversations episode
(http://www.blogtalkradio.com/biota/2008/06/28/the-Artificial-Life-of-Plants-1),
I got to meet Stefan Bornhofen, and hear about his plant simulation.
I would love to invite him up to talk about it, and I asked him at the
end of the episode.

http://www.math-info.univ-paris5.fr/alife/projects/Perelin/proj-bottom.htm

Maybe we could all chip in and pay his train fare or something.

--
Gerald de Jong
Beautiful Code BV
http://www.beautifulcode.eu
ph: +31629339805

[Anne van Rossum]

Hi Gerald, hi all,

Yes, would be awesome to have the next meeting. Last weeks I was busy
with implementing a wireless sensor network at my work. It uses a
light sensor on each so-called sensor mote, and the combination of all
those motes allows it to recognize patterns made by a torch. So, if
the network is attached to a wall, flashing the torch at the right can
mean: open the door; at the left can mean: close the door. This can be
used to detect suspicious movements in underground parking places e.g.
The cool thing about it from Greythumb perspective is probably that it
uses neural networks. A very simple version: a Kohonen network, but
still. :-)

My time in the States was great! As you could read in my last
newsletter, Josh Bongard, was there (http://www.cs.uvm.edu/
~jbongard/ ) presenting simulations in which bodies learn to move and
where the genes for morphology and locomotion are coevolved. On of his
interesting (IMHO) findings was, that mother evolution apparently
preferred to make a decomposition in genes coding mainly for
morphology and genes coding mainly for locomotion. It's was -
apparently again - beneficial to be able to make "evolutionary
mistakes" that only involved locomotion, or only involved form.

Next weeks I will try to tackle the SMART system. Spike-based ART
(Adaptive Resonance Theory), which uses top-down templates as well as
bottom-up templates in a neural network. This works as a very flexible
memory system. Depends on how far I come with this, but that's what I
would like to present most. It is a neural network that can be
implemented on the organisms in a simulated world and it provides them
with long(er) term memory. Perhaps not needed yet, but interesting
enough to hear something about I guess. :-)

Cheers!

Anne

On Aug 27, 9:07 pm, "Gerald de Jong" <geralddej...@gmail.com> wrote:
> I've been talking to Joost Rekveld (http://www.lumen.nu/rekveld/wp/)
> about a date for the second Greythumb NL gathering, and we're thinking
> about the last part of september, at his academie in Den Haag.
>
> I'd like to ask you to think about what you might present or talk
> about, if you've ever thought about standing up.  The bar isn't very
> high.  Even stories about a few evenings spent playing with an
> existing artificial life simulation would be great!  We can all learn
> from that.  This is supposed to be fun.
>
> I was hoping to hear from Anne van Rossum about his project, focusing
> on what is biologically inspired or derived.
>
> From a very interesting biota conversations episode

> (http://www.blogtalkradio.com/biota/2008/06/28/the-Artificial-Life-of-...),

> I got to meet Stefan Bornhofen, and hear about his plant simulation.
> I would love to invite him up to talk about it, and I asked him at the
> end of the episode.
>

> http://www.math-info.univ-paris5.fr/alife/projects/Perelin/proj-botto...

[Anne van Rossum]

Ah, I forgot. Sure, some money for a train ticket, no problem!

PS: Cool, Joost Rekveld's website!

On Aug 27, 9:07 pm, "Gerald de Jong" <geralddej...@gmail.com> wrote:

> I've been talking to Joost Rekveld (http://www.lumen.nu/rekveld/wp/)
> about a date for the second Greythumb NL gathering, and we're thinking
> about the last part of september, at his academie in Den Haag.
>
> I'd like to ask you to think about what you might present or talk
> about, if you've ever thought about standing up.  The bar isn't very
> high.  Even stories about a few evenings spent playing with an
> existing artificial life simulation would be great!  We can all learn
> from that.  This is supposed to be fun.
>
> I was hoping to hear from Anne van Rossum about his project, focusing
> on what is biologically inspired or derived.
>
> From a very interesting biota conversations episode

> (http://www.blogtalkradio.com/biota/2008/06/28/the-Artificial-Life-of-...),

> I got to meet Stefan Bornhofen, and hear about his plant simulation.
> I would love to invite him up to talk about it, and I asked him at the
> end of the episode.
>

> http://www.math-info.univ-paris5.fr/alife/projects/Perelin/proj-botto...

[Gerald de Jong]

So far it looks like it could be on the 23rd or the 25th of september.
So we can count on you to give a presentation, Anne?

Anybody else want to give a ten minute talk about some artificial life
program out there that they have played with??

--
Gerald de Jong

[Anne van Rossum]

Sure, I want to give a presentation. However, if you want to see
something simulated or a control architecture actually evolved, it is
better to wait for one month more I think. I will try to do my best to
have something for real tho. Cheers, Anne

[Gerald de Jong]

Great, thanks!

Nobody else for a ten-minute talk, get your feet wet?

--
Gerald de Jong