1. The A.I. Project
This is a multi-part message in MIME format.
------=_NextPart_000_0005_01C2857C.EC459FE0
charset="iso-8859-1"
Hi all,
This is such an interesting topic I have to put my 2 cents in.
Just to address the intelligence part of artificial intelligence.
Having all of the bodies physiological systems in place,I
think I would approach the brain's programming in two
parts,one part to control the body's physiology and the
other to create A CONSCIOUSNESS.
Like a neural net,our thoughts are all somehow relevant
to each other,creating kind of a steady stream of related
thoughts that continue from one to the next throughout=20
our entire waking lives, which in effect is what psychologists
call our consciousness.
With each new peice of information(variable),our consciousness seeks =
to find that new peice of information's
relevance to what we already know.
Even in cases of insanity,the subjects thoughts have been
proven to relate to one another,they dont just come from a
random place in the dark subconscious.
So I would establish a programmed loop of thoughts that
relate to each other and have one lead to the next. A small
loop to begin with,perhaps with some of the a.i's basic
drives.Then,as it learns new things,I would first have the=20
a.i sort it's relevance,then insert it somewhere in the loop
of its own consciousness.
I would think that we would also have to establish an
environment for the a.i.Each person is a product of
their environment,upbringing and experience. So where
does Darius exist,does it have an environment and how
would that affect the stream of its consciousness?
Anyway, I've got programming diagrams and notes on things like =
emotion and expression if anyones interested
in this approach.
We should continue this thread,perhaps all contribute
to Darius,or start from scratch.A euphoria community
a.i. project.What would you all think of this?
Or generally speaking,what would the community think
of any kind of group project? I think any organized project that =
consists of more than say ten people in
collaboration would be a feat in itself.
Most decent software today is the result of a good team.
We should get organized,do something...really create
something big...like an a.i. project!!=20
Thanks for listening
JDUBE
------=_NextPart_000_0005_01C2857C.EC459FE0
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META content=3D"text/html; charset=3Diso-8859-1" =
http-equiv=3DContent-Type>
<META content=3D"MSHTML 5.00.3315.2870" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT face=3DArial size=3D2>Hi all,</FONT></DIV>
<DIV> </DIV>
<DIV><FONT face=3DArial size=3D2> This is such an interesting topic =
I have to=20
put my 2 cents in.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2> Just to address the intelligence =
part of=20
artificial intelligence.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2> Having all of the bodies =
physiological=20
systems in place,I</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>think I would approach the brain's =
programming in=20
two</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>parts,one part to control the body's =
physiology and=20
the</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>other to create A =
CONSCIOUSNESS.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2> Like a neural net,our thoughts =
are all=20
somehow relevant</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>to each other,creating kind of a steady =
stream of=20
related</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>thoughts that continue from one to the =
next=20
throughout </FONT></DIV>
<DIV><FONT face=3DArial size=3D2>our entire waking lives, which in =
effect is what=20
psychologists</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>call our consciousness.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2> With each new peice of=20
information(variable),our consciousness seeks to find that new peice of=20
information's</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>relevance to what we already =
know.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2> Even in cases of insanity,the =
subjects=20
thoughts have been</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>proven to relate to one another,they =
dont just come=20
from a</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>random place in the dark =
subconscious.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2> So I would establish a =
programmed loop of=20
thoughts that</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>relate to each other and have one lead =
to the next.=20
A small</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>loop to begin with,perhaps with some of =
the a.i's=20
basic</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>drives.Then,as it learns new things,I =
would first=20
have the </FONT></DIV>
<DIV><FONT face=3DArial size=3D2>a.i sort it's relevance,then insert it =
somewhere in=20
the loop</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>of its own consciousness.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2> I would think that we would also =
have to=20
establish an</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>environment for the a.i.Each person is =
a product=20
of</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>their environment,upbringing and =
experience. So=20
where</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>does Darius exist,does it have an =
environment and=20
how</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>would that affect the stream of its=20
consciousness?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2> Anyway, I've got =
programming diagrams=20
and notes on things like emotion and expression if anyones=20
interested</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>in this approach.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2> We should continue this =
thread,perhaps=20
all contribute</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>to Darius,or start from scratch.A =
euphoria=20
community</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>a.i. project.What would you all think =
of=20
this?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2> Or generally speaking,what would =
the=20
community think</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>of any kind of group project? I think =
any organized=20
project</FONT><FONT face=3DArial size=3D2> that consists of more =
than say ten=20
people in</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>collaboration would be a feat in=20
itself.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2> Most decent software today is =
the result of=20
a good team.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>We should get organized,do =
something...really=20
create</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>something big...like an a.i.=20
project!!</FONT> </DIV>
<DIV> </DIV>
<DIV><FONT face=3DArial=20
size=3D2> &nbs=
p;  =
; =
=20
Thanks for listening</FONT></DIV>
<DIV><FONT face=3DArial=20
size=3D2> &nbs=
p;  =
; =
&=
nbsp; =20
------=_NextPart_000_0005_01C2857C.EC459FE0--
2. Re: The A.I. Project
I wouldn't mind helping sculpt an AI project, just to see what the many
brains of this EUPHORIA group can do.
I'm thinking we'd need to start with a worm, though. A worm is probably all
instinct, so it would give us a good foundation on composing the instinct of
a more advanced intelligence. Or maybe not. :)
Here's what our basis could be:
What We Need
----------------------
Virtual Environment
Virtual Worm with Sensory Functions (let's start him off with smell and
touch) and Instinctual Functions (i.e., survival).
Let me ask though... If we created an artificially intelligent worm, would
it eventually learn enough to break out of its mold and become a super
genius worm? Let's find out... :)
----- Original Message -----
From: <dubetyrant at hotmail.com>
To: "EUforum" <EUforum at topica.com>
Sent: Wednesday, November 06, 2002 9:11 AM
Subject: The A.I. Project
Hi all,
This is such an interesting topic I have to put my 2 cents in.
Just to address the intelligence part of artificial intelligence.
Having all of the bodies physiological systems in place,I
think I would approach the brain's programming in two
parts,one part to control the body's physiology and the
other to create A CONSCIOUSNESS.
Like a neural net,our thoughts are all somehow relevant
to each other,creating kind of a steady stream of related
thoughts that continue from one to the next throughout
our entire waking lives, which in effect is what psychologists
call our consciousness.
With each new peice of information(variable),our consciousness seeks to
find that new peice of information's
relevance to what we already know.
Even in cases of insanity,the subjects thoughts have been
proven to relate to one another,they dont just come from a
random place in the dark subconscious.
So I would establish a programmed loop of thoughts that
relate to each other and have one lead to the next. A small
loop to begin with,perhaps with some of the a.i's basic
drives.Then,as it learns new things,I would first have the
a.i sort it's relevance,then insert it somewhere in the loop
of its own consciousness.
I would think that we would also have to establish an
environment for the a.i.Each person is a product of
their environment,upbringing and experience. So where
does Darius exist,does it have an environment and how
would that affect the stream of its consciousness?
Anyway, I've got programming diagrams and notes on things like emotion
and expression if anyones interested
in this approach.
We should continue this thread,perhaps all contribute
to Darius,or start from scratch.A euphoria community
a.i. project.What would you all think of this?
Or generally speaking,what would the community think
of any kind of group project? I think any organized project that consists of
more than say ten people in
collaboration would be a feat in itself.
Most decent software today is the result of a good team.
We should get organized,do something...really create
something big...like an a.i. project!!
Thanks for listening
JDUBE
==^^===============================================================
This email was sent to: cklester at yahoo.com
3. Re: The A.I. Project
- Posted by tone.skoda at gmx.net
Nov 06, 2002
This is a multi-part message in MIME format.
------=_NextPart_000_000E_01C285C8.008BF1C0
charset="iso-8859-1"
Good idea for AI project is to have game like pacman, just a little =
adjusted. Enemies have health from -100 till +100. Enemies are =
identified with random numbers (colors). When pacman and some enemy meet =
pacman eats enemy, that means health of enemy is added to my man. Eaten =
enemy doesn't disappear, maybe just for a few moves after its been =
eaten. Health of enemy is only seen when pacman eats enemy. Pacman will =
never die, its health will just go up and down. High health is good, low =
is bad.=20
Senses: To simplify things it will only see (or touch) what's one unit =
in front of it, nothing else.
Movement: it will be able to move one unit forward and backwards and =
turn left and right. (4 possible actions)
First will move pacman, then all enemis, and so on. Like chess. One move =
at a time, to simplify things.
Here's a little hard to know how would AI pacman have to move to be =
intelligent. That could be solved that human plays pacman.
You wouldnt see the whole maze but only what's in front of you, so it =
will be either: wall, space or some enemy.
You will be allowed to write anything on paper, to figure out structure =
of maze for example, or movement of enemies.
Your movement will be recorded and you'll be able to watch later how you =
moved in maze from top perspective, to know what's intelligent and how =
would AI pacman have to move to be intelligent.
For the structure of AI brain I would have combination of some form of =
neural net and evolution algorithm, better forms of neural nets will be =
favored by this evolution algorithm.
Last time I gave you wrong link for C Elegans worm, here's more =
interesting link: http://glendhu.com/ai/neuroscience/wormbrain/
Here are some more of my thoughts:
Sensory information just triggers "pre-programmed"
actions in the brain. Everything in the brain
is pre-programmed. And it changes all the time
when we receive new sensory information.
For example, a tick will byte everything that has 37 C and
smells of butric acid. When that combination=20
(or sequence) of sensory information
goes into its brain it allways flows
thru same connections between neurons
towards motor neurons.
Two more things which support this theory that
everything in brain is preprogrammed:
a) the big number of neurons in the brain
and b) the speed by which the brain works.
----- Original Message -----=20
From: dubetyrant at hotmail.com=20
To: EUforum=20
Sent: Wednesday, November 06, 2002 4:11 PM
Subject: The A.I. Project
Hi all,
This is such an interesting topic I have to put my 2 cents in.
Just to address the intelligence part of artificial intelligence.
Having all of the bodies physiological systems in place,I
think I would approach the brain's programming in two
parts,one part to control the body's physiology and the
other to create A CONSCIOUSNESS.
Like a neural net,our thoughts are all somehow relevant
to each other,creating kind of a steady stream of related
thoughts that continue from one to the next throughout=20
our entire waking lives, which in effect is what psychologists
call our consciousness.
With each new peice of information(variable),our consciousness seeks =
to find that new peice of information's
relevance to what we already know.
Even in cases of insanity,the subjects thoughts have been
proven to relate to one another,they dont just come from a
random place in the dark subconscious.
So I would establish a programmed loop of thoughts that
relate to each other and have one lead to the next. A small
loop to begin with,perhaps with some of the a.i's basic
drives.Then,as it learns new things,I would first have the=20
a.i sort it's relevance,then insert it somewhere in the loop
of its own consciousness.
I would think that we would also have to establish an
environment for the a.i.Each person is a product of
their environment,upbringing and experience. So where
does Darius exist,does it have an environment and how
would that affect the stream of its consciousness?
Anyway, I've got programming diagrams and notes on things like =
emotion and expression if anyones interested
in this approach.
We should continue this thread,perhaps all contribute
to Darius,or start from scratch.A euphoria community
a.i. project.What would you all think of this?
Or generally speaking,what would the community think
of any kind of group project? I think any organized project that =
consists of more than say ten people in
collaboration would be a feat in itself.
Most decent software today is the result of a good team.
We should get organized,do something...really create
something big...like an a.i. project!!=20
Thanks for listening
JDUBE
------=_NextPart_000_000E_01C285C8.008BF1C0
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=3DContent-Type content=3D"text/html; =
charset=3Diso-8859-1">
<META content=3D"MSHTML 6.00.2600.0" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT face=3DArial size=3D2>Good idea for AI project is to have =
game like=20
pacman, just a little adjusted. Enemies have health from -100 =
till=20
+100. Enemies are identified with random numbers (colors). When =
pacman and=20
some enemy meet pacman eats enemy, that means health of enemy is added =
to my=20
man. Eaten enemy doesn't disappear, maybe just for a few moves after its =
been=20
eaten. Health of enemy is only seen when pacman eats enemy. =
Pacman=20
will never die, its health will just go up and down. High health is =
good, low is=20
bad. </FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Senses: To simplify things it will only =
see (or=20
touch) what's one unit in front of it, nothing else.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Movement: it will be able to move one =
unit forward=20
and backwards and turn left and right. (4 possible actions)</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>First will move pacman, then all =
enemis, and so on.=20
Like chess. One move at a time, to simplify things.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Here's a little hard to know how would =
AI pacman=20
have to move to be intelligent. That could be solved that human plays=20
pacman.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>You wouldnt see the whole maze but only =
what's in=20
front of you, so it will be either: wall, space or some =
enemy.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>You will be allowed to write anything =
on paper, to=20
figure out structure of maze for example, or movement of =
enemies.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Your movement will be recorded and =
you'll be able=20
to watch later how you moved in maze from top perspective, to know =
what's=20
intelligent and how would AI pacman have to move to be =
intelligent.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT> </DIV>
<DIV><FONT face=3DArial size=3D2>For the structure of AI brain I would =
have=20
combination of some form of neural net and evolution algorithm, better =
forms of=20
neural nets will be favored by this evolution algorithm.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT> </DIV>
<DIV><FONT face=3DArial size=3D2></FONT> </DIV>
<DIV><FONT face=3DArial size=3D2>Last time I gave you wrong link for C =
Elegans worm,=20
here's more interesting link: <A=20
href=3D"http://glendhu.com/ai/neuroscience/wormbrain/">http://glendhu.com=
/ai/neuroscience/wormbrain/</A></FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT> </DIV>
<DIV><FONT face=3DArial size=3D2>Here are some more of my =
thoughts:</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT> </DIV>
<DIV><FONT face=3DArial size=3D2> Sensory information just =
triggers=20
"pre-programmed"<BR> actions in the brain. Everything in the=20
brain<BR> is pre-programmed. And it changes all the time<BR> =
when we=20
receive new sensory information.<BR> For example, a tick will byte =
everything that has 37 C and<BR> smells of butric acid. When that=20
combination <BR> (or sequence) of sensory information<BR> =
goes into=20
its brain it allways flows<BR> thru same connections between=20
neurons<BR> towards motor neurons.<BR> Two more things which =
support=20
this theory that<BR> everything in brain is =
preprogrammed:<BR> a)=20
the big number of neurons in the brain<BR> and b) the speed by =
which the=20
brain works.</FONT></DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV style=3D"FONT: 10pt arial">----- Original Message ----- </DIV>
<DIV=20
style=3D"BACKGROUND: #e4e4e4; FONT: 10pt arial; font-color: =
black"><B>From:</B>=20
<A title=3Ddubetyrant at hotmail.com=20
href=3D"mailto:dubetyrant at hotmail.com">dubetyrant at hotmail.com</A> =
</DIV>
<DIV style=3D"FONT: 10pt arial"><B>To:</B> <A =
title=3DEUforum at topica.com=20
href=3D"mailto:EUforum at topica.com">EUforum</A> </DIV>
<DIV style=3D"FONT: 10pt arial"><B>Sent:</B> Wednesday, November 06, =
2002 4:11=20
PM</DIV>
<DIV style=3D"FONT: 10pt arial"><B>Subject:</B> The A.I. Project</DIV>
<DIV><BR></DIV><PRE>=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D The Euphoria =
Mailing List =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=20
</PRE>
<DIV><FONT face=3DArial size=3D2>Hi all,</FONT></DIV>
<DIV> </DIV>
<DIV><FONT face=3DArial size=3D2> This is such an interesting =
topic I have to=20
put my 2 cents in.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2> Just to address the =
intelligence part of=20
artificial intelligence.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2> Having all of the bodies =
physiological=20
systems in place,I</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>think I would approach the brain's =
programming in=20
two</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>parts,one part to control the body's =
physiology=20
and the</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>other to create A =
CONSCIOUSNESS.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2> Like a neural net,our thoughts =
are all=20
somehow relevant</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>to each other,creating kind of a =
steady stream of=20
related</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>thoughts that continue from one to =
the next=20
throughout </FONT></DIV>
<DIV><FONT face=3DArial size=3D2>our entire waking lives, which in =
effect is what=20
psychologists</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>call our consciousness.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2> With each new peice of=20
information(variable),our consciousness seeks to find that new peice =
of=20
information's</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>relevance to what we already =
know.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2> Even in cases of insanity,the =
subjects=20
thoughts have been</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>proven to relate to one another,they =
dont just=20
come from a</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>random place in the dark=20
subconscious.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2> So I would establish a =
programmed loop of=20
thoughts that</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>relate to each other and have one =
lead to the=20
next. A small</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>loop to begin with,perhaps with some =
of the a.i's=20
basic</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>drives.Then,as it learns new things,I =
would first=20
have the </FONT></DIV>
<DIV><FONT face=3DArial size=3D2>a.i sort it's relevance,then insert =
it somewhere=20
in the loop</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>of its own =
consciousness.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2> I would think that we would =
also have to=20
establish an</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>environment for the a.i.Each person =
is a product=20
of</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>their environment,upbringing and =
experience. So=20
where</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>does Darius exist,does it have an =
environment and=20
how</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>would that affect the stream of its=20
consciousness?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2> Anyway, I've got =
programming=20
diagrams and notes on things like emotion and expression if =
anyones=20
interested</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>in this approach.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2> We should continue this =
thread,perhaps=20
all contribute</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>to Darius,or start from scratch.A =
euphoria=20
community</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>a.i. project.What would you all think =
of=20
this?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2> Or generally speaking,what =
would the=20
community think</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>of any kind of group project? I think =
any=20
organized project</FONT><FONT face=3DArial size=3D2> that =
consists of more=20
than say ten people in</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>collaboration would be a feat in=20
itself.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2> Most decent software today is =
the result=20
of a good team.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>We should get organized,do =
something...really=20
create</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>something big...like an a.i.=20
project!!</FONT> </DIV>
<DIV> </DIV>
<DIV><FONT face=3DArial=20
=
size=3D2> &nbs=
p;  =
; =
=20
Thanks for listening</FONT></DIV>
<DIV><FONT face=3DArial=20
=
size=3D2> &nbs=
p;  =
; =
&=
nbsp; =20
=
JDUBE</FONT></DIV><PRE>=3D=3D^=3D=3D^=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
This email was sent to: tone.skoda at gmx.net
EASY UNSUBSCRIBE click here: <A =
href=3D"http://topica.com/u/?b1dd66.b3fcy3.dG9uZS5z">http://topica.com/u/=
?b1dd66.b3fcy3.dG9uZS5z</A>
Or send an email to: EUforum-unsubscribe at topica.com
T O P I C A -- Register now to manage your mail!
<A =
href=3D"http://www.topica.com/partner/tag02/register">http://www.topica.c=
om/partner/tag02/register</A>
=3D=3D^=3D=3D^=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
------=_NextPart_000_000E_01C285C8.008BF1C0--
4. Re: The A.I. Project
> Good idea for AI project is to have game like pacman, just a little
adjusted.
In your suggestion, do we create an AI Pacman or AI enemies?
> Here's a little hard to know how would AI pacman have to move to be
intelligent.
The point of AI is that YOU don't determine how it will move... You let it
learn how to move!
> That could be solved that human plays pacman.
How about simply "survival?" :)
> For the structure of AI brain I would have combination of
> some form of neural net and evolution algorithm, better forms
> of neural nets will be favored by this evolution algorithm.
The "Pacman in a Maze" suggestion is actually a pretty good one for a simple
AI test. One good thing is we're limiting its world to a simple maze, so we
don't have to worry about a lot of extraneous otherworldly stuff. So here's
my next Simple AI Entity Contest suggestion:
Develop a virtual worm that will live in a virtual glass bowl. Its world is
limited to the bowl.
> Sensory information just triggers "pre-programmed"
> actions in the brain. Everything in the brain
> is pre-programmed. And it changes all the time
> when we receive new sensory information.
That doesn't make sense, does it? It's either preprogrammed or it changes.
Which is it?
If it's pre-programmed, then what is learning? I propose learning is
developing a skill as opposed to "finding" it in one's skillset.
The only thing that is "pre-programmed" is instinct... The rest we gotta
learn.
> Two more things which support this theory that
> everything in brain is preprogrammed:
> a) the big number of neurons in the brain
> and b) the speed by which the brain works.
Take juggling for instance: The non-juggler cannot just pick up and go.
However, with training, new neurons are formed and their ability to juggle
not only develops, their proficiency gets better as they continue to
practice!
That's learning. Acquiring new skills (useful or not).
5. Re: The A.I. Project
- Posted by tone.skoda at gmx.net
Nov 06, 2002
----- Original Message -----
From: "C. K. Lester" <cklester at yahoo.com>
> In your suggestion, do we create an AI Pacman or AI enemies?
AI pacman. Enemies will just circle around in same simple or complicated
paths. Maybe later one
enemy
could be replaced with human player.
> Here's a little hard to know how would AI pacman have to move to be
> intelligent.
>
> The point of AI is that YOU don't determine how it will move... You let it
> learn how to move!
And I will, but you have to know easily when it's behaving intelligent.
> > That could be solved that human plays pacman.
>
> How about simply "survival?" :)
I wouldn't put those things in AI, survival, replication. It's not needed.
It played major role in biological development of intelligence, but AI
enviroment can be anything you want.
Survival is just natural selection, ie the evolution. It takes too much
time.
> Develop a virtual worm that will live in a virtual glass bowl. Its world
is
> limited to the bowl.
So when will you know when worm is behaving intelligent? In pacman maze its
much easier to see when its starts to behave intelligent. Worm also lives in
too simple world: it needs to swim towards food or away from toxic
chemicals. You could reach than behavior pretty soon without worm really
having some important intelligence.
> That doesn't make sense, does it? It's either preprogrammed or it changes.
> Which is it?
I don't mean pre-programmed like it's born with it. When we learn something
some connections change between neurons. When we will see same/similar thing
next time we will react same or similar, because most of the same neurons
and
connections between them will work.
6. Re: The A.I. Project
> > In your suggestion, do we create an AI Pacman or AI enemies?
>
> AI pacman. Enemies will just circle around in same simple or complicated
> paths. Maybe later one enemy could be replaced with human player.
If your Pacman is going to have "enemies," won't you necessarily need a
survival instinct? You say that your pacman will never die, but it will
get
weaker. What's to motivate it to get stronger?
> > The point of AI is that YOU don't determine how it will move... You
let
it
> > learn how to move!
>
> And I will, but you have to know easily when it's behaving intelligent.
A Pacman who learns to survive can't be that dumb. ;)
Plus, will you be measuring it on a scale?! I mean, what's the difference
between a Pacman with IQ of 20 and one with IQ of 120? And won't either
one
of them get bored eventually?
Remember, we're not talking about smart robots here, we're talking about
artificial intelligence, maybe even artificial life!
> I wouldn't put those things in AI, survival, replication. It's not
needed.
Replication, no; survival? yes! Otherwise, what's the point? :)
Now, since your "enemies" are really just poison pills, the "survival
instinct" of your Pacman will simply be a "maintain a high number"
instinct.
> Survival is just natural selection, ie the evolution.
I'm not talking about the survival of a species... I'm talking about the
survival instinct of an AI entity/individual.
> It takes too much time.
> > Develop a virtual worm that will live in a virtual glass bowl. Its
world
> > is limited to the bowl.
>
> So when will you know when worm is behaving intelligent?
> In pacman maze its much easier to see when its starts to behave
intelligent.
Not really. Besides, you're severely limiting your Pacman to see only one
square ahead! With no evolutionistic mechanism for improving vision in
subsequent generations, how will they ever get beyond guessing what's up
ahead? All I see for your Pacman is it remembering the pattern of the maze
and maybe locations of bite-sized healthy bits.
7. Re: The A.I. Project
- Posted by tone.skoda at gmx.net
Nov 06, 2002
----- Original Message -----
From: "C. K. Lester" <cklester at yahoo.com>
> If your Pacman is going to have "enemies," won't you necessarily need a
> survival instinct? You say that your pacman will never die, but it will
> get
> weaker. What's to motivate it to get stronger?
I would replace survival with "good-bad". All in its world could be divided
between good and bad things: good and bad actions (or sequence or
combination of actions) it takes.
Sensory information will, in some way, sculpture its neural net. In what way
needs to be found out yet.
We need to find if there is something in common how different sensory
information sculptures neural net.
That thing that is in common to learning is present in all intelligent
creatures.
> A Pacman who learns to survive can't be that dumb. ;)
>
> Plus, will you be measuring it on a scale?! I mean, what's the difference
> between a Pacman with IQ of 20 and one with IQ of 120? And won't either
> one
> of them get bored eventually?
If finding food and avoiding poison is most important thing in their life
then they wont get bored.
I'm not sure if feelings are really needed in AI and if AI creature will
have them.
> Remember, we're not talking about smart robots here, we're talking about
> artificial intelligence, maybe even artificial life!
I can't just make AI creature and world and enemies and leave it running all
night and then next morning I will have real AI :) I have to be able to
expolore what types of algorithms, neural nets etc are good.
> Now, since your "enemies" are really just poison pills, the "survival
> instinct" of your Pacman will simply be a "maintain a high number"
> instinct.
What's wrong with that? I want to achieve intelligence, I'm not interested
in what will AI creature feel, what will be its purpose in life, survival or
pills. Those "pills" are very important to it.
> Not really. Besides, you're severely limiting your Pacman to see only one
> square ahead!
For simplicity. It can still do a lot with this vision :) I want to be able
to actually see how neural net is structured and that's why I need simple
input. Later I will move on to different enviroment ofcourse, like 3d game.
You have to start small. :)
There is a lot of complexity already here if you know that it will have to
remember all the sequence of actions it takes in its neural net and if they
were good or bad.
>All I see for your Pacman is it remembering the pattern of the maze
> and maybe locations of bite-sized healthy bits.
Remembering pattern is quite important.
8. Re: The A.I. Project
On 0, tone.skoda at gmx.net wrote:
> Here are some more of my thoughts:
>
> Sensory information just triggers "pre-programmed"
> actions in the brain. Everything in the brain
> is pre-programmed. And it changes all the time
> when we receive new sensory information.
No, Very little human intellegence is preprogrammed in the brain
(however lower animal instinct and such is, i think this is what you
meant).
> For example, a tick will byte everything that has 37 C and
> smells of butric acid. When that combination
> (or sequence) of sensory information
> goes into its brain it allways flows
> thru same connections between neurons
> towards motor neurons.
True, but thats instinct, not intellegence.
(Our most advanced AIs tend to mimic instict rather than intellegence,
however.)
> Two more things which support this theory that
> everything in brain is preprogrammed:
> a) the big number of neurons in the brain
> and b) the speed by which the brain works.
a) and b) are related, and are not such proof, each neuron is a simple
processor, the large number lets parallel processing run things much
faster
than if only a few neurons (small scale SMP) or 1 neuron (single
processor)
was doing them.
By the way, I was wrong. You can theoreticly have a human level
intellegence
on a PC if you had enough memory to store all the information the AI
needed and were willing to settle on speed. (BTW the human brain may
store its
information in as little space as a few megabytes (not including motor
neurons and the like) and I'd be suprised if it were more than a few TB
altogether, though few PCs have that much hard disk space.) Remember that
its
possible for a single processor to emulate multiple processors, at the
sacrafice of speed and memory.
jbrown
--
9. Re: The A.I. Project
This is all nice and good, but its really AI: Artifical Instinct.
Such a simple game wont give AI. (I dare say that all but the most
advanced
chess programs are also mainly instinct.) Not that thats a bad thing,
advanced enough instict will still get lots of things done, and instict
to
mimic human intellegence has been done and is very realistic, if it is
advanced enough.
jbrown
On 0, tone.skoda at gmx.net wrote:
>
>
> ----- Original Message -----
> From: "C. K. Lester" <cklester at yahoo.com>
>
> > If your Pacman is going to have "enemies," won't you necessarily need a
> > survival instinct? You say that your pacman will never die, but it will
> > get
> > weaker. What's to motivate it to get stronger?
>
> I would replace survival with "good-bad". All in its world could be divided
> between good and bad things: good and bad actions (or sequence or
> combination of actions) it takes.
> Sensory information will, in some way, sculpture its neural net. In what way
> needs to be found out yet.
> We need to find if there is something in common how different sensory
> information sculptures neural net.
> That thing that is in common to learning is present in all intelligent
> creatures.
>
>
> > A Pacman who learns to survive can't be that dumb. ;)
> >
> > Plus, will you be measuring it on a scale?! I mean, what's the difference
> > between a Pacman with IQ of 20 and one with IQ of 120? And won't either
> > one
> > of them get bored eventually?
>
> If finding food and avoiding poison is most important thing in their life
> then they wont get bored.
> I'm not sure if feelings are really needed in AI and if AI creature will
> have them.
>
> > Remember, we're not talking about smart robots here, we're talking about
> > artificial intelligence, maybe even artificial life!
>
> I can't just make AI creature and world and enemies and leave it running all
> night and then next morning I will have real AI :) I have to be able to
> expolore what types of algorithms, neural nets etc are good.
>
>
> > Now, since your "enemies" are really just poison pills, the "survival
> > instinct" of your Pacman will simply be a "maintain a high number"
> > instinct.
>
> What's wrong with that? I want to achieve intelligence, I'm not interested
> in what will AI creature feel, what will be its purpose in life, survival or
> pills. Those "pills" are very important to it.
>
>
> > Not really. Besides, you're severely limiting your Pacman to see only one
> > square ahead!
>
> For simplicity. It can still do a lot with this vision :) I want to be able
> to actually see how neural net is structured and that's why I need simple
> input. Later I will move on to different enviroment ofcourse, like 3d game.
> You have to start small. :)
> There is a lot of complexity already here if you know that it will have to
> remember all the sequence of actions it takes in its neural net and if they
> were good or bad.
>
> >All I see for your Pacman is it remembering the pattern of the maze
> > and maybe locations of bite-sized healthy bits.
>
> Remembering pattern is quite important.
>
>
>
>
--
10. Re: The A.I. Project
Yes! This is a good idea. Simplifes amtters a lot. Don't have fancy
hardware?
Emulate it with graphics routines. Can't use the graphics interface? Then
use the text one! The more portable it is between inferfaces, the more
advanced it can be while the more people can work on it.
jbrown
On 0, jbrown105 at speedymail.org wrote:
>
> On 0, tone.skoda at gmx.net wrote:
> > Here are some more of my thoughts:
> >
> > Sensory information just triggers "pre-programmed"
> > actions in the brain. Everything in the brain
> > is pre-programmed. And it changes all the time
> > when we receive new sensory information.
>
> No, Very little human intellegence is preprogrammed in the brain
> (however lower animal instinct and such is, i think this is what you
> meant).
>
> > For example, a tick will byte everything that has 37 C and
> > smells of butric acid. When that combination
> > (or sequence) of sensory information
> > goes into its brain it allways flows
> > thru same connections between neurons
> > towards motor neurons.
>
> True, but thats instinct, not intellegence.
> (Our most advanced AIs tend to mimic instict rather than intellegence,
> however.)
>
> > Two more things which support this theory that
> > everything in brain is preprogrammed:
> > a) the big number of neurons in the brain
> > and b) the speed by which the brain works.
>
> a) and b) are related, and are not such proof, each neuron is a simple
> processor, the large number lets parallel processing run things much
> faster
> than if only a few neurons (small scale SMP) or 1 neuron (single
> processor)
> was doing them.
>
> By the way, I was wrong. You can theoreticly have a human level
> intellegence
> on a PC if you had enough memory to store all the information the AI
> needed and were willing to settle on speed. (BTW the human brain may
> store its
> information in as little space as a few megabytes (not including motor
> neurons and the like) and I'd be suprised if it were more than a few TB
> altogether, though few PCs have that much hard disk space.) Remember that
> its
> possible for a single processor to emulate multiple processors, at the
> sacrafice of speed and memory.
>
> jbrown
>
>
> --
>
>
>
>
Linux User:190064
Linux Machine:84163
--
11. Re: The A.I. Project
> I've been sifting through the a.i thoughts weve gotten so far and one
> thing that seems persistent in what you have been arguing is the
simulation
> vs. emulation arguments, or perhaps better termed pre-programmed vs.
instinct
> only programming.
Here's the distinction between the two concepts:
1. simulation - the AI entity exists in a virtual world, getting its input
from and providing output to that virtual world. It knows nothing of the
outside world. This would include our worm in a glass bowl scenario. We can
observe and interact with the AI entity. The key here is that the
environment is simulated!
2. actual - the AI entity exists as code on a PC. Its environment is the PC.
It can obtain input the way the PC obtains input (desktop cam, scanner,
internet, etc.) and can output as the PC can output (printers, screen,
posting to a web site, etc.). The key here is that the environment is
actual.
> I need to know how your
> theories would play out put in a neural net programming perspective.
In both cases, a neural net can function as the AI entity's brain.
> Would this "core instinct programming" approach mesh well with
> neural nets.In other words would your a.i. programming approach
> involve the building of neural nets?
Absolutely. First you need to understand the neural net's capabilities.
Then, you'll want to program a "wrapper" to it- this would be the instinct
layer. The instinct layer would include all instinctual and autonomic
functions.
> In one context,the use of the prime motivator,"or survival instinct",
> was to be depicted as a number,like 10,if we use neural nets,we'll not
> be able to program attributes like the survival instinct in the form
> of numbers,but rather as a series of built up neurons which together
> form a neural network.
Yes, you can use static numbers. The key is, the AI entity won't understand
the mechanism anyway.
Look at human reactions: we know there is a fight or flight response
mechanism, but we can't explain it! What motivates our brain to signal
"fear" when our lives are in jeopardy? What motivates our brain to CARE
about survival?! Our brain functions without us having to understand how it
functions. This might be the problem... unless we can understand how the
brain operates, we'll never be able to create our own creature.
So, what I'm saying is, you could have the survival instinct simply be a
measurement on a scale...
> How do you think we should go at neural network
> programming in the light of say,building up a prime motivator?
Intelligence is built upon instinct.
Instinct is built upon a brain.
The structure of the brain will determine how intelligent the creature will
get.
We have amazingly complex brains which will probably never be successfully
understood or emulated.
>From my observations, humans are highly complex (and that's an
understatement) chemical computers. Worms, on the other hand, along with
most (if not all) of the animal kingdom, are highly complex chemical
*robots* or expert systems. What component do we possess that lets us
achieve sentience/intelligence? Answer that and you might have something...
:)
12. Re: The A.I. Project
- Posted by tone.skoda at gmx.net
Nov 07, 2002
----- Original Message -----
From: "C. K. Lester" <cklester at yahoo.com>
> No problem- "maintain a high number" is just a less-fatal version of
> "avoid death." I guess I was thinking that "fear of death" would be a
> great motivator.
For biological creatures avoiding death is sure main motivator, but for AI
creature living in simulated enviroment it can be anything. It's not
important at all what it is. Later if
Ai creature will be put into real world it will ofcourse need survival
motive.
> But what if those feelings are necessary for intelligence? I hate to
> keep repeating myself, but "fear of death" might be REQUIRED just to get
> the creature functioning to survive.
Why would it have to worry to survive if it won't be able to die? :)
I would look at AI this way: sensory information is very important to
achieve intelligence.
Sensory information forms AI brain.
If we look at the nature there is a lot of different kind of visual, audio,
chemical etc sensory information available to be processed by the brains,
any brains, not just human. So the brains are just very well adjusted to the
type of sensory information that they receive. You could look at sensory
information as a very large data which looks random at first, but it's not
and it has some complicated pattern. We're just figuring pattern from
sensory information.
We should just concentrate on how the sensory information forms neural net
and ignore other thoughts what should AI creature feel and what should its
motive be.
One more important thing about sensory information: complication of it
really comes of SEQUENCE and COMBINATION of sensory information we recieve.
To me discussions what are simulations, what are expert systems and what is
true AI are not really leading anywhere. If you are looking at it that way
then all AI available today are just expert systems. The border is not
clear.
Discussions about feelings, consciousnes and such things also seem boring to
me. They are maybe some side effect of the brain, but main task of brain is
to choose right decision/action among many available. Brain constantly tunes
itself to be able to cope with complex and changing enviroment around it.
That's why brain was invented by evolution :)
Our brain gets just the right balance of fixed intelligence from our genes
as it's neccesary. If it was more our bran would be less adjustable.
Evolution would be able to produce realy intelligent "expert system" brain
which would be born intelligent and would do only one thing really good and
have less abilities to change its structure. But it wouldn't be able to
adjust to new things and that creature would die.
I wonder how would look brain of human which would have no senses right from
birth:
no vision, hearing, touch, smell, taste.
Good approach would be if sensory information would in different random ways
form connections in neural net.Then the best way by which neural net is
modified from sensory information
is selected with evolution algorithm.
13. Re: The A.I. Project
- Posted by dm31 at uow.edu.au
Nov 07, 2002
Well. Anyways, I'll repeat what I said at first.
We need a GOOD NN in EU first, if will will be using one in anything
along these longs.
NN's are more then just nodes with weighted inputs. That is the basic
defination of a NN, but misses alot when I comes to making a useful NN
Just doing this would be a big enough challange at the moment. We
would need to think about such this as whether would would create a
Feed-forward (associative) network or a Feedback (autoassociative)
network. Then would we do single layer NN's or support the more
complex multilayer systems.
For the transfer function would we use either linear (or ramp),
threshold or sigmoid??
Since we would most likey use a Adaptive network over a fixed network
judging on what we are talking about would you use the
Back-Propagation Algorithm (most widely used) for adjusting the
weights as it learns?
Will we have it setup for Supervised or unsupervised training?
(yes CK, check it up, those are the two ways to classify an Adaptive
network :>)
These are questions and problems we most solve first. Lets not get to
ahead of our selves at the start, it would be a shame if we started
something like what ppl are proposing and it falls apart because we
didn't think it through at the start :( I would love to see as build a
project along some of the lines been suggested, lets just get the
software we would need first.
Cheers,
Dan
14. Re: The A.I. Project
The last is a neural networks lib.
http://www.rapideuphoria.com/burgess.zip
http://www.rapideuphoria.com/genlib.zip
http://www.rapideuphoria.com/nnet.zip
=====
Best Regards,
Guillermo Bonvehi
AKA: Knixeur - Caballero Rojo
--- dm31 at uow.edu.au wrote:
>
> Well. Anyways, I'll repeat what I said at first.
>
> We need a GOOD NN in EU first, if will will be using one in anything
> along these longs.
>
> NN's are more then just nodes with weighted inputs. That is the basic
> defination of a NN, but misses alot when I comes to making a useful
> NN
>
> Just doing this would be a big enough challange at the moment. We
> would need to think about such this as whether would would create a
> Feed-forward (associative) network or a Feedback (autoassociative)
> network. Then would we do single layer NN's or support the more
> complex multilayer systems.
>
> For the transfer function would we use either linear (or ramp),
> threshold or sigmoid??
>
> Since we would most likey use a Adaptive network over a fixed network
> judging on what we are talking about would you use the
> Back-Propagation Algorithm (most widely used) for adjusting the
> weights as it learns?
>
> Will we have it setup for Supervised or unsupervised training?
> (yes CK, check it up, those are the two ways to classify an Adaptive
> network :>)
>
> These are questions and problems we most solve first. Lets not get to
> ahead of our selves at the start, it would be a shame if we started
> something like what ppl are proposing and it falls apart because we
> didn't think it through at the start :( I would love to see as build
> a
> project along some of the lines been suggested, lets just get the
> software we would need first.
>
> Cheers,
> Dan
>
> ==^^===============================================================
> This email was sent to: pampeano at rocketmail.com
>
>
>
15. Re: The A.I. Project
- Posted by dm31 at uow.edu.au
Nov 07, 2002
|http://www.rapideuphoria.com/burgess.zip
|http://www.rapideuphoria.com/genlib.zip
|http://www.rapideuphoria.com/nnet.zip
|
|=====
|Best Regards,
| Guillermo Bonvehi
| AKA: Knixeur - Caballero Rojo
I know. I have had them for ages. infact, I have downloaded alot of
the eu libs before, and have them on backup cd.
In my opinion(note, only mine), this is ok for a basic back-p network,
but is not a GOOD NN(not refering to the guys programming, just the
implemention).
Thats why I said we should make one first, that can handle what we
needed. Complex NN algrothim's are relatively 'longish' and complex
bits of code. I have seen some of the C++ code up in the reseach labs
here, and it is pages long. I'm gunna see if they will let me use
parts of the code and wrap as a dll for EU.
16. Re: The A.I. Project
On 0, Rom <kjehas at frisurf.no> wrote:
> Some thoughts about Jdube's original message.
>
> First Jdube seems to assume that an ordinary programming language is
> suitable for logical (AI) programming. I disagree. I think he is a bit
> optimistic here.
>
It depends on what part of the program you consider to b the "brain" of
the AI. If you mean the Euphoria statements and such, then yes its not
adaquete. However, if you mean the sequence of data used for the genetic
algorithem or the neural net or whatever, then it is adaquete, although
technically the AI is the sequence, and not the Eu code.
> > ... so I would establish a programmed loop of thoughts....
>
> That is easy to program. Remember that the CPU and Windows are running
> loops all the time. These loops are are sending messages ... if that
> counts? This is not exactly what we usually term as "consciousness".
>
The problem comes with defining consciousness. I cant say that such a
loop
would make the computer consious, but I would say that it would give it
the ability to "think", in a limited sense. (Again, though, this depends
on how you define "think".)
> > ... then as i learns new things....
>
>
> Introducing learning comes a bit fast here.
>
> What is learning? If learning is to update data in some database, then
> learning is easy to program. I would suggest that a large program that
> is learning must have some ability to change itself .... producing
> program code that it can execute.
>
What if the database itself _is_ the "program code"? (Of course, that
would
mean that Euphoria only manifests the AI's brain, but I dont remember
anyone who argued that this could be done in EU that the EU statements
_themselves_ were to be the brain of the AI.)
> So already at the very start of this discussion I disagree.
>
> Euphoria isn't suitable for logical programming. But Euphoria could
> probably be made suitable (if one is optimistic) ....
>
> 1) A small interpretor like Euphoria can be control by another program.
> Source can be generated, loaded and executed. If the program halts? ...
> well, then it had to be aborted (not posssible today ... I don't know).
>
> 2) Euphoria is so small today that it can be changed to support new
> functionality ..... too late to do the same with C.
>
> 3) For a program to be changeable I suppose the program syntax has to be
> very simple. Euphoria is maybe the simplest language that is....still
> powerful. The simpler program code is... the more realistic it will be
> for a program to change the code.
>
> The simplest program code I can think about is rules (if A the B). The
> simplicity of rules is such that it doesn't matter if the rules are put
> into a database, not the program itself. Rules can be evaluated in that
> case too.
>
> Conclusion.
> Trying to simulate worms or things like that will fail because Euphoria
> hasn't been adapted to logical programming. Those who think this is just
> a matter of ordinary programming cannot even create a decent demo expert
> system with Euphoria ... I am afraid.
Ok, you're right. This is not a matter of ordinary programming. But keep
in mind that languages such a Prolog and LISP are written in another
language (C i believe, I remember seeing one implementation of LISP in
Java ...) and that with the proper programming it is most certainly
possible in EU.
>
> Rom
>
jbrown
--
17. Re: The A.I. Project
On 9 Nov 2002, at 7:05, jbrown105 at speedymail.org wrote:
>
> On 0, Rom <kjehas at frisurf.no> wrote:
<snip>
> > Conclusion.
> > Trying to simulate worms or things like that will fail because Euphoria
> > hasn't been adapted to logical programming. Those who think this is just a
> > matter of ordinary programming cannot even create a decent demo expert
> > system
> > with Euphoria ... I am afraid.
>
> Ok, you're right. This is not a matter of ordinary programming. But keep
> in mind that languages such a Prolog and LISP are written in another
> language (C i believe, I remember seeing one implementation of LISP in
> Java ...) and that with the proper programming it is most certainly
> possible in EU.
But the Eu interpreter is also written in C, like the early interpreters for the
"recognised" Ai languages. Adding a few features to Eu will result in the
same abilities. Wasn't Google written in Lisp?
Kat
18. Re: The A.I. Project
This is a multi-part message in MIME format.
------=_NextPart_000_0057_01C287F9.58758B20
charset="iso-8859-1"
"Kat"
>But the Eu interpreter is also written in C, like the early =
interpreters for the "recognised" Ai languages. Adding a few features to =
Eu will result in the same abilities.
Easy for me to accept.
I imaging a message-handling system is something that is problably =
required when complexity of a program reaches that of a newer game. This =
is something that does not have to be reinvented, just copied from =
existing solutions. Event-handling (Win32Lib) is message-handling.
I also imaging a new data type for representing "rules" .... Eu =
"objects" does not have a clear semantic meaning. "Objects" with =
associative power may do the same job as nodes in a neural network are =
supposed to do? Some general format is needed so that the system always =
know how to handle these "objects", and knows how to compare one =
"object" with another "object". The meaning og these "objects" is =
releated to the messages that passes through the system .... or related =
to other objects... nothing more.
I say nothing else that if a problem is to be solved inside the Eu =
community, then the Eu language must have the power to handle the =
problem. Otherwise the problem is beyond the reach of the Eu community.
Rom
------=_NextPart_000_0057_01C287F9.58758B20
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=3DContent-Type content=3D"text/html; =
charset=3Diso-8859-1">
<META content=3D"MSHTML 6.00.2600.0" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT face=3DArial size=3D2>
<P><FONT face=3D"Times New Roman" size=3D3>"Kat"</FONT></P>
<P><FONT face=3D"Times New Roman" size=3D3>>But the Eu interpreter is =
also=20
written in C, like the early interpreters for the "recognised" Ai =
languages.=20
Adding a few features to Eu will result in the same =
abilities.</FONT></P>
<P><FONT face=3D"Times New Roman" size=3D3>Easy for me to =
accept.</FONT></P>
<P><FONT face=3D"Times New Roman" size=3D3>I imaging a message-handling =
system is=20
something that is problably required when complexity of a program =
reaches that=20
of a newer game. This is something that does not have to be reinvented, =
just=20
copied from existing solutions. Event-handling (Win32Lib) is=20
message-handling.</FONT></P>
<P><FONT face=3D"Times New Roman" size=3D3>I also imaging a new data =
type for=20
representing "rules" .... Eu "objects" does not have a clear semantic =
meaning.=20
"Objects" with associative power may do the same job as nodes in a =
neural=20
network are supposed to do? Some general format is needed so that the =
system=20
always know how to handle these "objects", and knows how to compare one =
"object"=20
with another "object". The meaning og these "objects" is releated to the =
messages that passes through the system .... or related to other =
objects...=20
nothing more.<BR>I say nothing else that if a problem is to be solved =
inside the=20
Eu community, then the Eu language must have the power to handle the =
problem.=20
Otherwise the problem is beyond the reach of the Eu =
community.</FONT></P>
<P><FONT face=3D"Times New Roman" =
------=_NextPart_000_0057_01C287F9.58758B20--
19. Re: The A.I. Project
Hi all,
>I say nothing else that if a problem is to be solved inside the Eu
>community, then the Eu language must have the power to handle the problem.
>Otherwise the problem is beyond the reach of the Eu community.
>
>Rom
Very true.But you have to know what your programming before you can
worry about how to program it(speaking for myself)I have only a
general knowledge of NN, I've been looking into projects of this
nature that have already been done by others examining how they approached
programming.It doesn't take long to realize there is alot
of material out there,thanks for some great links so
far,Turing,Creatures,some archive examples(though apparently Darius is
top secret),even one or two science fiction authors.
> Good Nite
JDUBE
>==^^===============================================================
>This email was sent to: dubetyrant at hotmail.com
>
>
20. Re: The A.I. Project
This is a multi-part message in MIME format.
------=_NextPart_000_0010_01C28919.722AF200
charset="iso-8859-1"
I have done surfing on=20
http://www.alicebot.org/=20
Very amusing to read. Some lines copied below.
Rom
(from The Anatomy of A.L.I.C.E. by Dr. Richard S. Wallace)
A general downturn in artificial intelligence and robotics roughly =
coincided with the end of the Cold War, as governments and corporations =
reduced the amount of funding available for this technology. The "richly =
funded" field of 1976 became more like a Darwinian struggle for =
diminishing resources. One positive outcome was the brief heyday of =
"robot minimalism," a design philosophy based on low-cost parts, =
commodity computers, low-bandwidth sensing, and general simplicity in =
design and engineering. It was a moment when Occam's razor could cut =
away much of the needless complexity that had accumulated over the =
previous decades. Although robot minimalism subsequently fell out of =
favor, it became a significant influence on the development of =
A.L.I.C.E.=20
We used to say there was no theory behind A.L.I.C.E., no neural =
networks, no knowledge representation, no deep search, no genetic =
algorithms and no parsing. Then we discovered that there was a theory =
circulating in applied A.I., called Case-Based Reasoning (CBR) [CBR?? =
reference] that closely resembled the stimulus-response structure of =
A.L.I.C.E. The CBR cases correspond to the AIML categories.
------=_NextPart_000_0010_01C28919.722AF200
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=3DContent-Type content=3D"text/html; =
charset=3Diso-8859-1">
<META content=3D"MSHTML 6.00.2600.0" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT face=3DArial size=3D2><FONT size=3D4>
<P><FONT face=3D"Times New Roman" size=3D3>I have done surfing on =
</FONT></P>
<P><FONT face=3D"Times New Roman" size=3D3>http://www.alicebot.org/ =
</FONT></P>
<P><FONT face=3D"Times New Roman" size=3D3>Very amusing to read. Some =
lines copied=20
below.</FONT></P>
<P><FONT face=3D"Times New Roman" size=3D3>Rom</FONT></P>
<P><FONT size=3D2></FONT> </P></FONT><FONT size=3D4>
<P><FONT face=3D"Times New Roman"><STRONG>(</STRONG>from <STRONG>The =
Anatomy of=20
A.L.I.C.E.</STRONG></FONT><FONT face=3D"Times New Roman"> by Dr. Richard =
S.=20
Wallace)</FONT></P></FONT><FONT size=3D4></FONT><FONT face=3D"Times New =
Roman"=20
size=3D4>
<P align=3Djustify>A general downturn in artificial intelligence and =
robotics=20
roughly coincided with the end of the Cold War, as governments and =
corporations=20
reduced the amount of funding available for this technology. The "richly =
funded"=20
field of 1976 became more like a Darwinian struggle for diminishing =
resources.=20
One positive outcome was the brief heyday of "robot minimalism," a =
design=20
philosophy based on low-cost parts, commodity computers, low-bandwidth =
sensing,=20
and general simplicity in design and engineering. It was a moment when =
Occam=92s=20
razor could cut away much of the needless complexity that had =
accumulated over=20
the previous decades. Although robot minimalism subsequently fell out of =
favor,=20
it became a significant influence on the development of A.L.I.C.E. </P>
<P>We used to say there was <I>no</I> theory behind A.L.I.C.E., no =
neural=20
networks, no knowledge representation, no deep search, no genetic =
algorithms and=20
no parsing. Then we discovered that there was a theory circulating in =
applied=20
A.I., called Case-Based Reasoning (CBR) [CBR?? reference] that closely =
resembled=20
the stimulus-response structure of A.L.I.C.E. The CBR cases correspond =
to the=20
AIML categories.</P></FONT><FONT size=3D2>
------=_NextPart_000_0010_01C28919.722AF200--
21. Re: The A.I. Project
- Posted by tone.skoda at gmx.net
Nov 10, 2002
I have put up a new site with some interesting artificial intelligence
links:
http://www10.brinkster.com/tskoda/AILinks.htm
22. Re: The A.I. Project
On 11 Nov 2002, at 0:29, Rom wrote:
>
> I have done surfing on
>
> http://www.alicebot.org/
>
> Very amusing to read. Some lines copied below.
>
> Rom
>
>
> (from The Anatomy of A.L.I.C.E. by Dr. Richard S. Wallace)
>
> A general downturn in artificial intelligence and robotics roughly coincided
> with the end of the Cold War, as governments and corporations reduced the
> amount
> of funding available for this technology. The "richly funded" field of 1976
> became more like a Darwinian struggle for diminishing resources. One positive
> outcome was the brief heyday of "robot minimalism," a design philosophy based
> on
> low-cost parts, commodity computers, low-bandwidth sensing, and general
> simplicity in design and engineering. It was a moment when Occam's razor could
> cut away much of the needless complexity that had accumulated over the
> previous
> decades. Although robot minimalism subsequently fell out of favor, it became a
> significant influence on the development of A.L.I.C.E.
>
> We used to say there was no theory behind A.L.I.C.E., no neural networks, no
> knowledge representation, no deep search, no genetic algorithms and no
> parsing.
> Then we discovered that there was a theory circulating in applied A.I., called
> Case-Based Reasoning (CBR) [CBR?? reference] that closely resembled the
> stimulus-response structure of A.L.I.C.E. The CBR cases correspond to the AIML
> categories.
Also referred to as scripts. In humans, we call that "real life experience". You
know when you eat in a resturant, you are eating the food you ordered. The
computer never went to a resturant, so you haveto spell out that info
explicitly in some form, even if it's explicit in an indirect way over several
"sentences" or "paragraphs". Many Ai projects have 100,000's people-hours
in programming things like that. Many others, like Cyc, have million of hours
(over 10 yrs) in detailing the items like food(resturant(etc(etc))). I
downloaded
and spent months studying Cyc source data tables.
Kat
23. Re: The A.I. Project
----- Original Message -----=20
From: "Kat" <kat at kogeijin.com>
> > We used to say there was no theory behind A.L.I.C.E., no neural =
networks, no
> > knowledge representation, no deep search, no genetic algorithms and =
no parsing.
> > Then we discovered that there was a theory circulating in applied =
A.I., called
> > Case-Based Reasoning (CBR) [CBR?? reference] that closely resembled =
the
> > stimulus-response structure of A.L.I.C.E. The CBR cases correspond =
to the AIML
> > categories.
>=20
>
> Also referred to as scripts.=20
Probably I don't disagree with you. Just became curious why you are =
saying "script" in this context.=20
Is "Cyc" based on scripts? I don't know. Is Cyc based on CBR too?
Rom
24. Re: The A.I. Project
On 11 Nov 2002, at 2:53, Rom wrote:
>
> ----- Original Message -----
> From: "Kat" <kat at kogeijin.com>
>
> > > We used to say there was no theory behind A.L.I.C.E., no neural networks,
> > > no
> > > knowledge representation, no deep search, no genetic algorithms and no
> > > parsing. Then we discovered that there was a theory circulating in applied
> > > A.I., called Case-Based Reasoning (CBR) [CBR?? reference] that closely
> > > resembled the stimulus-response structure of A.L.I.C.E. The CBR cases
> > > correspond to the AIML categories.
> >
> >
> > Also referred to as scripts.
>
> Probably I don't disagree with you. Just became curious why you are saying
> "script" in this context. Is "Cyc" based on scripts? I don't know. Is Cyc
> based
> on CBR too?
Scripts were the term used way back when, they outlined cases of behavior,
or events, or paths thru a timeline. The script in this context wasn't an
executeable. some were very basic, like "if x is on y, then y is under x",
which seems to be an easy executeable in "real Ai languages" (not Eu), and
testable in reverse by a function call. The biggest problem i have with scripts
is the assumption by the computer that if A and C-F happened, then B
happened too. Strangely, this was accepted behavior, even tho it is illogical
and smells like a lawyer leading the witness. I don't remember if scripts were
installed in Cyc or not. They were looking for "common sense" to "happen"
once enough real-world data was fed in, but it never happened. Cyc is built
on 1000's of these things, 100's of which were released, and 1000's which
are secret:
#$objectMoving : <#$MovementEvent> <#$PartiallyTangible>
The predicate #$objectMoving is used to refer to an object which is moving in
a particular #$MovementEvent. (#$objectMoving MOVE OBJECT) means that
OBJECT is in motion at some point during the #$MovementEvent MOVE and
this movement is focal in MOVE. In some cases, such as an instance of
#$LocomotionEvent, OBJECT causes MOVE directly. OBJECT's motion in
MOVE may be translational, rotational, or some combination of these.
OBJECT may be a solid -- such as an animal, an automobile, or a tennis ball
-- or a fluid, such as lava flowing or a warm air mass rising.
isa: #$ActorSlot #$AsymmetricBinaryPredicate #$IrreflexiveBinaryPredicate
genlPreds: #$transferredThing #$temporallyIntersects
some more specialized predicates: #$objectEmitted #$objectRemoved
#$transporter #$transportees #$vehicle #$mainTransportees #$driverActor
#$passengers
Just remembering the names of the events, and removing duplicates, must
have been a huge time-wasting job. As far as i remember, there were no
examples like the CBR or scripts built into Cyc, the last time i took a peek
into it's insides.
At least one program with CBR and scripts was able to munge the scripts,
and write short tv programs in the 70's.
Kat
25. Re: The A.I. Project
> > > > We used to say there was no theory behind A.L.I.C.E., no neural =
networks, ...
The chatterbot A.L.I.C.E is basicly a terrible small program (50 pages I =
think). It has been translated from Java to PHP, Perl, Delphi (could not =
download that version) and C++. If fact, Euphoria/EDS could have been =
the perfect choice (but no one has done that).
The PandoraBox team are anticiapating ALICE as a common frontend =
program.... forget about Windows?
The ALICE community is maybe the same size as Euphoria, however, I think =
they have something that will attract more attention .... even if ALICE =
itself is much simpler than Eu. Any reflections?
I have worked for some time with case-based-reasoning (CBR). My =
conclusion about CBR is that it as uncomplicated as ALICE and very =
powerful. Given an ALICE-like tool, able to read "rules" in XML format =
(....mostly what users have to learn), marking up a medical textbook is =
as easy as to read one, and result is an instant diagnosing expert.
Euphoria, ALICE and CBR have something in common...simplicity.
Rom
26. Re: The A.I. Project
Could you state which specs you'd expect for an adequate NN lib, or
give us some reference to such specs?
Regards
CChris
> Date: Fri, 08 Nov 2002 18:38:00 +1100
> From: dm31 at uow.edu.au
> Subject: Re: The A.I. Project
>
>
> |http://www.rapideuphoria.com/burgess.zip
> |http://www.rapideuphoria.com/genlib.zip
> |http://www.rapideuphoria.com/nnet.zip
> |
> |=====
> |Best Regards,
> | Guillermo Bonvehi
> | AKA: Knixeur - Caballero Rojo
>
> I know. I have had them for ages. infact, I have downloaded alot of
> the eu libs before, and have them on backup cd.
>
> In my opinion(note, only mine), this is ok for a basic back-p network,
> but is not a GOOD NN(not refering to the guys programming, just the
> implemention).
>
> Thats why I said we should make one first, that can handle what we
> needed. Complex NN algrothim's are relatively 'longish' and complex
> bits of code. I have seen some of the C++ code up in the reseach labs
> here, and it is pages long. I'm gunna see if they will let me use
> parts of the code and wrap as a dll for EU.
>
>
