DPRG
DPRG List  



DPRG: Re: Little Ricci 2

Subject: DPRG: Re: Little Ricci 2
From: Ed Koffeman edk at cyberramp.net
Date: Wed Jan 7 12:01:38 CST 1998

to S Sunder:

Your English is very good.

Any member can jump in and add their thoughts on the subject at hand to the
discussions on this mailing list.  That is exactly what is supposed to
happen here.

You appear to have clearly stated some basic principles that should not be
forgotten.  Thank you.

The robot that was shown at our meeting a while back was impressive in the
intelligence of its behavior, but that was really an example of the
intelligence of its creator rather than a real-time intelligence created
through experience.  Its algorithms were carefully crafted through
intelligent thought and trial, then embedded into the unit.

This does not make the achievement any less, because we were all very
impressed with what it could do.  But it is still an example of great skill
(the ability to move about an uncontrolled environment without getting stuck
or damaged) rather than machine 'intelligence'.

I have heard of one definition of insanity, being that an insane person will
continue to do perform damaging or fruitless activity, when a presumably
intelligent person will quickly modify their behavior.  Presumably the
intelligent person quickly gains enough data to create the knowledge
required to predict an unsuccessful outcome if the same behavior is
repeated.

So it would seem to me you need data (from sensors, communication etc.), an
ability to analyze the data to extract meaning from it (thinking), and a
method of deciding whether or how to modify its behavior using the extracted
meaning.  If the behaviour changes, it will presumably have 'learned' from
its experiences.

If it makes bad decisions about how to modify its behavior (which must
likely be put to the test before the outcome is seen to be better or worse)
then it will still seem not to be intelligent.  Perhaps this is the fine
line between high intelligence and insanity.  Robert Heinlein said "a
sufficiently advanced technology is indistinguishable from magic", and
(Koffeman's Law?) in some cases it may be that a sufficiently advanced
intelligence is indistinguishable from insanity<grin>.  There are probably a
lot of cases (wartime?) when people do seemingly idiotic things until the
true purpose becomes clear, and then it is seen to be a highly intelligent
thing to have done.  (Of course the observer usually does not have the same
data to work with.)


Ed Koffeman


- -----Original Message-----
>From: S Sunder <sunder at rocketmail.com>
To: dprglist at dprg.org <dprglist at dprg.org>
Date: Wednesday, January 07, 1998 8:11 AM
Subject: Re: DPRG: Re: Little Ricci 2


>Hi Friends!
>
>I looked at few mails asking questions about the
>Intelligence in the Robots.
>
>Well, an intellegent system (it could be a robot) learns and
>uses information (knowledge) on its own. The actions
>of the systems are as per the character of that system
>which is altered only by the knowledge (past experiences)
>and its basic understanding of the current environment.
>
>It is also possible that an Intelligent system can
>work along with other intellegent systems but not fully
>govern by the external system.
>
>If you have a DOG and a DOG like toy pulled by a child.
>Which do you think will behave in intellegent way? They
>may go in same direction but the actions of the DOG are
>more under its own making.
>
>While the DOG has a lower degree of intellegence it
>known its master and others and uses its knowledge
>in its actions. It will also learn from sitiation.
>
>The lower degree of intellegence is rather intutive
>or simply a response to the signals but there is hardly
>any knowledge. Insects go by senses and random motion.
>No, knowledge. If the senses become dull ot the response
>of the senses is modified then and only then the actions
>may differ in such cases.
>
>Remote controlled toys are not intelligent as long as
>they can not learn from the experiences and act on their
>own.
>
>It is not very important on how you introduce the
>intelligence in the system at the first place. It should
>be there in one form or the other.
>
>If all senses are dead and there are actions then they
>are only intutive of the system. They are not intelligent
>actions. Response to the past wold of knowledge and the
>current world of experience must govern the future. That
>is intelligence.
>
>I was not to answer this question yet I did so please
>forgive me for causing an interference. Did you get
>anything out of it at the first place? My English ??!!
>
>Shyam India
>
>
>===
>      ________  ________  _________   ________
>     /       / /       / /           /
>    /_______/ /       / /_______    /______
>   /  \      /       /         /   /
>  /    \    /_______/ ________/   /_______
>        "mailto:sunder at rocketmail.com"
>   "http://www.geocities.com/Eureka/7432/"
>
>
>
>
>---Michael Renzmann <renzman at stud.fh-frankfurt.de> wrote:
>>
>> Hi Ron.
>>
>> A few lines at the start: I read the thread between you and some others
>> in this
>> mailing-list. The information you gave sounded interesting, but I´m also
>> sceptic with new techniques. Especially with such I never heared before
>> except
>> one single source.
>> Please, don´t understand my questions as a force against you or your
>> technique.
>> But I have to ask some questions in order to make my own opinion of it.
>>
>> > Was it controlled by a radio or computer program?  If so it was not
>> > intelligent it only gave the appearance of intelligence.
>>
>> After reading this line I would be interested in your definition of
>> ´intelligence´.
>> But: please try to keep it as easy as possible to you, because I´m not
>> very keen in
>> english language and have some difficulties in understanding extrem
>> technical or
>> scientific words. Thanks.
>>
>> Bye, Mike
>>
>>
>>
>>
>
>_________________________________________________________
>DO YOU YAHOO!?
>Get your free @yahoo.com address at http://mail.yahoo.com
>
>
>

------------------------------

More information about the DPRG mailing list