Copy Link
Add to Bookmark
Report
AIList Digest Volume 7 Issue 049
AIList Digest Friday, 1 Jul 1988 Volume 7 : Issue 49
Today's Topics:
ZAD random number generator
Philosophy:
Who else isn't a science?
Reproducing the brain in low-power analog CMOS
replicating the brain with a Turing machine
linguistic metaphor for knowledge representation
more on dance notation
----------------------------------------------------------------------
Date: Sat, 25 Jun 88 00:58 O
From: <YLIKOSKI%FINFUN.BITNET@MITVMA.MIT.EDU>
Subject: ZAD random number generator
Distribution-File:
AILIST@AI.AI.MIT.EDU
In a recent AIList issue, it was inquired what kind of random number
generator a Zener diode / A-D converter generator would be.
I recall that noise from a Zener diode is quantum mechanical and
follows a well-known and well-defined theoretical spectrum (might it
be the 1/f law?). As is well known, distributions can be transformed
to obtain the desired one (exponential / Gaussian etc.).
Combining the results by Santha and Vazira mentioned by Albert
Boulanger with a quantum electrodynamical source of noise might even
best a good pseudorandom number generator.
Andy Ylikoski
------------------------------
Date: 27 Jun 88 00:18:24 GMT
From: bc@media-lab.media.mit.edu (bill coderre and his pets)
Subject: Re: Who else isn't a science?
In article <11387@agate.BERKELEY.EDU> weemba@garnet.berkeley.edu writes:
>Anyway, let me recommend the following works by neurophysiologists:
(references)
>These researchers start by looking at *real* brains, *real* EEGs, they
>work with what is known about *real* biological systems, and derive very
>intriguing connectionist-like models. To me, *this* is science.
And working in the other direction is not SCIENCE? Oh please...
>CAS & WJF have developed a rudimentary chaotic model based on the study
>of olfactory bulb EEGs in rabbits. They hooked together actual ODEs with
>actual parameters that describe actual rabbit brains, and get chaotic EEG
>like results.
There is still much that is not understood about how neurons work.
Practically nothing is known about how structures of neurons work. In
50 years, maybe we will have a better idea. In the mean time,
modelling incomplete and incorrect physical data is risky at best. In
the mean time, synthesizing models is just as useful.
>------------------------------------------------------------------------
>In article <2618@mit-amt.MEDIA.MIT.EDU>, bc@mit-amt (bill coderre) writes:
>>Oh boy. Just wonderful. We have people who have never done AI arguing
>>about whether or not it is a science [...]
>We've also got what I think a lot of people who've never studied the
>philosophy of science here too. Join the crowd.
I took a course from Kuhn. Speak for youself, chum.
>>May I also inform the above participants that a MAJORITY of AI
>>research is centered around some of the following:
>>[a list of topics]
>Which sure sounded like programming/engineering to me.
Oh excuse me. They're not SCIENCE. Oh my. Well, we can't go studying
THAT.
>> As it happens, I am doing simulations of animal
>>behavior using Society of Mind theories. So I do lots of learning and
>>knowledge acquisition.
>Well good for you! But are you doing SCIENCE? As in:
>If your simulations have only the slightest relevance to ethology, is your
>advisor going to tell you to chuck everything and try again? I doubt it.
So sorry to disappoint you. My coworkers and I are modelling real,
observable behavior, drawn from fish and ants. We have colleagues at
the New England Aquarium and Harvard (Bert Holldobler).
Marvin Minsky, our advisor, warns that we should not get "stuck" in
closely reproducing behavior, much as it makes no sense for us to
model the chemistry of the organism to implement its behavior (and
considering that ants are almost entirely smell-driven, this is not a
trite statement!).
The bottom line is that it is unimportant for us to argue whether or
not this or that is Real Science (TM).
What is important is for us to create new knowledge either
analytically (which you endorse) OR SYNTHETICALLY (which is just as
much SCIENCE as the other).
Just go ask Kuhn..........................................mr bc
heart full of salsa jalapena
------------------------------
Date: Wed, 29 Jun 88 08:23:28 PDT
From: John B. Nagle <jbn@glacier.stanford.edu>
Subject: Reproducing the brain in low-power analog CMOS
Forget Turing machines. The smart money is on reproducing the brain
with low-pwer analog CMOS VLSI. Carver Mead is down at Caltech, reverse
engineering the visual system of monkeys and building equivalent electronic
circuits. Progress seems to be rapid. Very possibly, traditional AI will
be bypassed by the VLSI people.
John Nagle
------------------------------
Date: Wed, 29 Jun 88 9:26:50 PDT
From: jlevy.pa@Xerox.COM
Subject: Re: AIList Digest V7 #46 replicating the brain with a
Turing machine
Andy Ylikoski asks why you can't replicate the brain's exact functions
with a Turing machine. First off, the brain is not a single machine but
a whole bunch of them. Therefore "replacing it with a Turing machine"
wouldn't get you there.
Turing machines have an inherent limitation in that they are not
reactive i.e. they are unable to react to the environment directly. On
the other hand, the brain is in direct communication with a number of
input devices (eyes, ears, nose, touch-sense, etc.), all of which are
sending data at the same time.
An interesting question is whether the brain's software suffers from the
Church-Rosser problem which is present in functional languages -
basically, you cannot, in a functional language, see that a certain
source of input is empty and later detect input on it. It seems that
this is not so, since we are able to close our eyes and later open them,
seeing again.
Just speculating...
--Jacob
References
AIList-REQUEST@AI.AI.MIT.EDU's message of Tue, 28 Jun 88 23:05:00 EDT
-- AIList Digest V7 #46
------------------------------
Date: Mon, 27 Jun 88 08:20:08 PDT
From: Stephen Smoliar <smoliar@vaxa.isi.edu>
Subject: linguistic metaphor for knowledge representation
Seth at BBN proposed a view of TYPES as "adjectives" and INSTANCES as "nouns,"
as an alternative to my view of TYPES as "nouns" (and INSTANCES sort of swept
under the rug as "entities"). Seth has a good point which, as I understand it,
actually has its roots in KL-ONE. (I suspect Ron Brachman will correct me if
my understanding is off.) The "types" of KL-ONE are called "classes;" and
while they might seem nominal, their usage is closer to adjectival. Thus,
while there might be a class called DOCTOR, the class is meant to embody the
DESCRIPTION of a doctor. I have heard various KL-ONE users employ phrases
like "doctor-like" or "doctor-ish" in discussing such a class. The adjectival
point of view becomes more apparent when you consider that a concept like
RICH-DOCTOR may be defined as a specialization of DOCTOR. This would perhaps
best be paraphrased as "having properties of being both doctor-like and rich."
Thus, there may be some merit in viewing classes as adjectival rather than
nominal.
------------------------------
Date: 27 Jun 88 18:57:13 GMT
From: dan@ads.com (Dan Shapiro)
Reply-to: dan@ads.com (Dan Shapiro)
Subject: Re: more on dance notation
As a method of preserving choreography, Labanotation has been close to
a complete failure; it is laborious, noncompact, nonvisual, and as
some people have commented, it doesn't capture relations of dancers in
space or to one another very well. As a result, few people are
skilled in writing or reading Labanotation, and the effort of
recording dances has been invested for only a very small percentage of
the world's choreographies. I have never heard of it being used to
generate choreography. For communication with people, video formats
are far more expressive.
As a candidate machine format, Labanotation has still more problems
which haven't been mentioned. It turns out that dance notation is not
only meant to capture the physical position of joints and the movement
of a dancer in space, but also the quality of the movement in an
emotional sense. Is the effort quick, percussive, or fluid, etc.?
Sometimes, this sense of the movement is more important than the joint
positions themselves. The formal vocabulary for recording these
qualities is quite limited - which means that dance notation is an
incomplete specification (dance performances are interesting
because there are as many ways of projecting a choreography as there
are dancers). However, it is unclear what the critical emotional
subset of choreography is. When would a choreographer be satisfied
that the recording is right?
My suspicion is that the most natural approach is to expand the
concept of dance notation to include both the static score, and an
interpreter which models the dancers that render the score visually.
These "dancers" would add their own nuances of interpretation.
There is a curious point here about the medium carrying the message.
Brings back memories of the 70s, doesn't it?
Dan Shapiro
------------------------------
Date: 28 Jun 88 09:52 PDT
From: hayes.pa@Xerox.COM
Subject: Re: AIList Digest V7 #45
On dance notation:
A quick suggestion for a similar but perhaps even thornier problem: a notation
for the movements involved in deaf sign language.
------------------------------
End of AIList Digest
********************
