Alt-BEAM Archive

Message #02996



To: "Dennison" dennlill@buffnet.net, "BEAM Mail" beam@corp.sgi.com
From: "Nigel Joyce" nigel.joyce@qr.com.au
Date: Thu, 6 May 1999 09:35:36 +1000
Subject: [alt-beam] Re: Multiple Inputs for Nv neurons



content-transfer-encoding: quoted-printable
content-type: text/plain;

I know it is mind numbing, that is why I only offered the Idea, not the sol=
ution. If it was easy, Mark T would have made the connection years ago. (=
That too is an assumption, I have never met him)
Blue Skies
---Buzz---
indeed what you say has merit. The idea of huge loops is an advantage of =
NV tech. The problem is, who's building the huge loops? Trying to get coord=
inated behavior out of large scale nv projects is mind numbing at the very =
least. =

=

Dennison =


OK this is going to be hard to explain but bear with me. I think that ro=
b is on the right path, concider the multitude of loops in a brain, I think=
that the interconnection between each loop could depend upon outside influ=
ences such as nerve endings. Like the different parts of the brain control=
ling different parts of the body and reacting to the inputs from those part=
s, lets assume (big, big, huge assumption) that in the brain, 1 Nv loop is=
attached to 1 nerve ending on the tip of a finger, you touch something, th=
e nerve ending acts like a PNC and starts the loop, the output from the loo=
p/chain goes somewhere else and tells another loop/chain something eg. the =
loop attached to a finger muscle telling it to stop the finger moving forwa=
rd. Now all we have to do is take the same approach and use 1 loop/chain p=
er output/feedback and have the feedback act as a PNC to other loops. Do y=
ou get what I mean?? It makes sense in my head. =


Not bad huh, I have been into beam for a month and made the breakthrough =
between real life and Alife {;-) =


Save a copy of this e-mail, as it will make me famous, ohHoHohahaha (mean=
t to be a mad scientist laugh) =


Either that or I am just starting to understand what Tilden is on about. =
{;-) =


Cya later =


Blue Skies
---Buzz---

I didn't just kiss death, I slipped it the tongue! =


nigel.joyce@usa.net
=


>Also, they need a form of regenerative feedback, or all the processes si=
mply end. The loop does that. The thing >that you may miss in a web like Nv=
, is that there are many, multiple loops made up by all the webs. You >coul=
dn't begin to count the massive loops in the human brain!!! That's quite th=
e daunting task to do in even an >earth work or cockroach even! Loops seem =
to be the best Central Pattern Generator,a nd they can branch off >into cha=
ins, rejoin, form other loops and do countless amazing things. Unfortunatel=
y, BEAM is in it's infancy, >and neural network based robots are well less =
than 100 years old (despite the fact that the neural net was >essentialy on=
e of the first "control" devices, created by nature itself!). Neural nets h=
ave had a while to develop, >and we are only trying to begin to catch up! =



Richard Piotter wrote: =

A web... Try looking at the interconnections in lobster. Even tho=
ugh there are two loops, they are attached in a "webby" layout, and that al=
one is cool (even if Mark T himself said Lobster realy was a disapointment =
performance wise). I don't necesarily use loops. MECI IV M and Spyder both =
use chains of Nvs. I have a lot of other ideas, but have not actualy done a=
nything. Spyder is my curent project. I'll be working on MECI IV M next, an=
d then maybe throw some turbot and head action into the mix! :) =

George Rix wrote: =


=




2) Why are microcores always connected in a loop, instead of in=
, say, a web of sorts, with plenty of interconnections? =

Signing off, =

Rob Rix 'If anything can possibly go wrong, it probably alrea=
dy has'-Murphy's law =

=

-- =





Richard Piotter =

richfil-@rconnect.com =



The Richfiles Robotics & TI web page: =

http://richfiles.calc.org =



For the BEAM Robotics list: =

BEAM Robotics Tek FAQ http://people.ne.mediaone.net/bushbo/beam/FAQ.html =





------------------------------------------------------------------------
eGroup home: http://www.eGroups.com/group/alt-beam
http://www.eGroups.com - Simplifying group communications


content-transfer-encoding: quoted-printable
content-type: text/html;



DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
x; PADDING-LEFT: 5px; PADDING-RIGHT: 0px">
indeed what you say has merit. T=
he idea =

of huge loops is an advantage of NV tech. The problem is, who's building =
the =

huge loops? Trying to get coordinated behavior out of large scale nv proj=
ects =

is mind numbing at the very least.

 

Dennison
NT>

OK this is going to be hard to explain but bear with me.  I think=
that =

rob is on the right path, concider the multitude of loops in a brain, I t=
hink =

that the interconnection between each loop could depend upon outside =

influences such as nerve endings.  Like the different parts of the b=
rain =

controlling different parts of the body and reacting to the inputs from t=
hose =

parts,  lets assume (big, big, huge assumption) that in the brain, 1=
Nv =

loop is attached to 1 nerve ending on the tip of a finger, you touch =

something, the nerve ending acts like a PNC and starts the loop, the=
=

output from the loop/chain goes somewhere else and tells another loop/cha=
in =

something eg. the loop attached to a finger muscle telling it to stop the=
=

finger moving forward.  Now all we have to do is take the same appro=
ach =

and use 1 loop/chain per output/feedback and have the feedback act as a P=
NC to =

other loops.  Do you get what I mean??  It makes sense in my he=
ad. =

Not bad huh, I have been into beam for a month and made the breakthrou=
gh =

between real life and Alife {;-) =

Save a copy of this e-mail, as it will make me famous, ohHoHohahaha (m=
eant =

to be a mad scientist laugh) =

Either that or I am just starting to understand what Tilden is on =

about.  {;-) =

Cya later =

Blue Skies
---Buzz---

 

I didn't just kiss death, I slipped it the tongue! =


 


href=3D"mailto:nigel.joyce@usa.net">nigel.joyce@usa.net

=

>Also, they need a form of regenerative feedback, or all the proces=
ses =

simply end. The loop does that. The thing >that you may miss in a web =
like =

Nv, is that there are many, multiple loops made up by all the webs. You =

>couldn't begin to count the massive loops in the human brain!!! That'=
s =

quite the daunting task to do in even an >earth work or cockroach even=
! =

Loops seem to be the best Central Pattern Generator,a nd they can branch =
off =

>into chains, rejoin, form other loops and do countless amazing things=
. =

Unfortunately, BEAM is in it's infancy, >and neural network based robo=
ts =

are well less than 100 years old (despite the fact that the neural net wa=
s =

>essentialy one of the first "control" devices, created by nature itse=
lf!). =

Neural nets have had a while to develop, >and we are only trying to be=
gin =

to catch up! =


Richard Piotter wrote:





  A web... Try looking at the interconnections in=
=

lobster. Even though there are two loops, they are attached in a "w=
ebby" =

layout, and that alone is cool (even if Mark T himself said Lobster=
=

realy was a disapointment performance wise). I don't necesarily use=
=

loops. MECI IV M and Spyder both use chains of Nvs. I have a lot of=
=

other ideas, but have not actualy done anything. Spyder is my curen=
t =

project. I'll be working on MECI IV M next, and then maybe throw so=
me =

turbot and head action into the mix! :) =

George Rix wrote:








 




BLE>-- =

  2) Why are microcores always connected in=
a =

loop, instead of in, say, a web of sorts, with plenty of =

interconnections? =

Signing off,
Rob Rix 'If anything can possibly go wron=
g, it =

probably already has'-Murphy's law




Richard Piotter

href=3D"http://www.egroups.com/person?protectID=3D19417621912202113213803=
8203148062063000026057218043130252055210">richfil-@rconnect.com =



The Richfiles Robotics & TI web page:

href=3D"http://richfiles.calc.org/" target=3D_top>http://richfiles.calc.o=
rg =


For the BEAM Robotics list:
BEAM Robotics Tek FAQ
href=3D"http://people.ne.mediaone.net/bushbo/beam/FAQ.html" =

target=3D_top>http://people.ne.mediaone.net/bushbo/beam/FAQ.html =


 





eGroup home: http://www.e=">http://www.eGroups.com/group/alt-beam">http://www.e=
Groups.com/group/alt-beam

www.eGroups.com">http://www.eGroups.com">www.eGroups.com - Simplifying group =
communications



Home