Copy Link
Add to Bookmark
Report
Neuron Digest Volume 08 Number 13
Neuron Digest Monday, 9 Dec 1991 Volume 8 : Issue 13
Today's Topics:
Backprop-simulator for Connection Machine available
half-baked notion
Request for ftp address
A robot called Frank.
Smalltalk Neural Network Program
Frank, the robot (?). The gathered information.
Postdoctoral work at Caltech
Cognitive/Experimental Psychologist position available
Cognitive Science at Birmingham
connectionism and social psychology
reminder and solicitation of INNS vote
job ad for NIPS message board
Graduate and Post-doc positions in Neural Processes in Cognition
Invariant pattern recognition with ANNs
Send submissions, questions, address maintenance, and requests for old
issues to "neuron-request@cattell.psych.upenn.edu". The ftp archives are
available from cattell.psych.upenn.edu (128.91.2.173). Back issues
requested by mail will eventually be sent, but may take a while.
----------------------------------------------------------------------
Subject: Backprop-simulator for Connection Machine available
From: Sebastian.Thrun@B.GP.CS.CMU.EDU
Date: Thu, 14 Nov 91 11:35:00 -0500
Folks,
according to the recent discussion concerning parallel implementations of
neural networks, I want to inform you that there is a very fast code of
backprop for the Connection Machine CM-2 public available. This posting
is about a year old, and meanwhile various labs have found and removed
(almost) all the bugs in the code, such that it is very reliable now. It
should be noted that this implementation is the most simple way of
parallelizing backprop (it's data-parallel), and that it works
efficiently with large training sets only (in the order of 500 to
infinity). But then it works great!
--- Sebastian
=------------------------------------------------------------------
(original message follows)
=------------------------------------------------------------------
The following might be interesting for everybody who works with the PDP
backpropagation simulator and has access to a Connection Machine:
********************************************************
** **
** PDP-Backpropagation on the Connection Machine **
** **
********************************************************
For testing our new Connection Machine CM/2 I extended the PDP
backpropagation simulator by Rumelhart, McClelland et al. with a parallel
training procedure for the Connection Machine (Interface C/Paris, Version
5).
Following some ideas by R.M. Faber and A. Singer I simply made use of the
inherent parallelism of the training set: Each processor on the
connection machine (there are at most 65536) evaluates the forward and
backward propagation phase for one training pattern only. Thus the whole
training set is evaluated in parallel and the training time does not
depend on the size of this set any longer. Especially at large training
sets this reduces the training time greatly. For example:
I trained a network with 28 nodes, 133 links and 23 biases to approximate
the differential equations for the pole balancing task adopted from
Anderson's dissertation. With a training set of 16384 patterns, using
the conventional "strain" command, one learning epoch took about 110.6
seconds on a SUN 4/110 - the connection machine with this SUN on the
frontend managed the same in 0.076 seconds.
--> This reduces one week exhaustive training to approximately seven
minutes!
(By parallelizing the networks themselves similar acceleration can be
achieved also with smaller training sets.)
=--------------
The source is written in C (Interface to Connection Machine: PARIS) and
can easily be embedded into the PDP software package. All origin
functions of the simulator are not touched - it is also still possible to
use the extended version without a Connection Machine. If you want to
have the source, please mail me!
Sebastian Thrun, thrun@cs.cmu.edu
You can also obtain the source via ftp:
ftp 129.26.1.90
Name: anonymous
Password: <transmit your full e-mail address, e.g. thrun@damn&hell.edu>
ftp> cd pub
ftp> cd gmd
ftp> get pdp-cm.c
ftp> bye
------------------------------
Subject: half-baked notion
From: "Raymond D. Scanlon" (CCB-R) <rscanlon@PICA.ARMY.MIL>
Date: Fri, 15 Nov 91 08:41:59 -0500
[[ Editor's Note: Half-baked? Perhaps. Provocative nonetheless. Consider
some recent work (Koch and ???) which implicated a 40 Hz EEG oscillation
as an external marker for "awareness" in cats and humans. I'll get the
citation and a short synopsis for next issue. -PM ]]
Alan Turing and John Searle revisited.
Turing's Test
Turing's test dates from 1950. In simple, the interrogator attempts to
figure out the sex of one or more hidden contestants. He can write
questions and read the answers. At least one contestant will lie with
intent to deceive. For the Test we replace a liar with a computer, and
"sex" with "is a human." If we can detect no difference in the game, we
say the computer is thinking.
Searle's Chinese Room
Searle's rejoinder dates from 1980. He invoked a Chinese Room. You, the
reader, are to imagine yourself in the room. There is an in-slot and an
out-slot. Through the in-slot come Chinese characters. You manipulate the
characters (called squiggles) by referring to a rule book and put them in
the out-slot. The rule book is to be so all inclusive that the outside
world understands a native Chinese speaker to inhabit the room. But you
know no Chinese at all. Searle says that a manipulative algebra and a
data base need not involve thinking.
The discussion is current. Scientific American (January 1990) contained
two articles on the subject. Presently in Boston, Turing's Test is
underway. Money is there.
We have misdirected our energies; it is not thinking but awareness we are
seeking. Of course a machine thinks; thinking is a function of the brain.
The question is not properly posed. We should ask "Is the machine aware?"
Alertness is objective; awareness is subjective.
Thinking and words have always been popular with people who deal in
words. Man is the only animal that talks. Man is the only rational
animal. Man is the only animal that thinks. There has always been a
desire to show that man is different from other animals.
I am aware.
You are alert.
He exhibits intelligent behavior.
Among vertebrates we extend awareness to all mammals. Among invertebrates
the octopus and the squid are a special problem. The giant squid has a
brain ten times the size of man's, but it is not a vertebrate brain. The
squid's genes built it differently. It might as well be from Mars. Is it
aware? To acknowledge awareness in one animal but deny it to another is
an act of conceit.
Awareness is a quality of the mind. We are aware of our thoughts, as we
are aware of existence, of sensations, and of our surroundings. Turing
was mislead, first by the emphasis on thinking and second by the notion
that thinking is the answering of questions. Question and answer is his
notion of how we display our intelligence. Why not let the machine ask
the questions--or better yet, let the machine interrupt us in
mid-sentence and finish the thought better than we could.
I say it is more reasonable, less egocentric to say consciousness has two
aspects, alert and aware, and forget thinking.
Why is there such a fuss?
The construction of machine brains belongs to neuroscientists, electrical
engineers, and computer programmers. Some people (philosophers and
mathematicians) feel left out. They respond with words and theorems and a
great fuss.
Does Turing's test give us any reason to think the machine is aware? No.
Thinking and the awareness of thinking exist in different universes.
Does Searle's Chinese Room give us any reason to think the machine is not
aware? No. Put yourself back in the Chinese Room. You know no Chinese but
you are aware that you are referring to a rule book and manipulating
characters. Similarly the machine might be fully aware that it is
manipulating pulses and sending certain ones to the video display.
Since the answer to both these questions is no, we see that neither
Turing nor Searle has anything to offer us. Can there be a test for
awareness? I think not. We confer the state of awareness; we do not test
it.
Ray Scanlon
Benet Laboratory
Watervliet Arsenal
Watervliet, NY 12189
rscanlon@pica.army.mil
------------------------------
Subject: Request for ftp address
From: "Frances Blomeley, King's College London" <UDAA032@oak.cc.kcl.ac.uk>
Date: Mon, 18 Nov 91 10:11:00 +0000
[[ Editor's Note: While the ftp archive on cattell has the original
UNIX/DOS code for the PDP software, I don't know where teh Mac version
is. I'd also like to get the "updated" code since I understand there were
a number of bug fixes. If anyone knows, please send me a note at
neuron-request@cattell.psych.upenn.edu. -PM ]]
I have lost the ftp address of the site holding the Rumelhart/McClelland
PDP software (for the Macintosh), and would be grateful if someone could
provide me with the information.
Many thanks
Frances Blomeley
Computing Centre
King's College London
------------------------------
Subject: A robot called Frank.
From: Paulo V Rocha <P.Rocha@cs.ucl.ac.uk>
Date: Tue, 19 Nov 91 14:43:22 +0000
The BBC Television (England) presented a programme on Neural Networks
where it featured, among many other things, a robot that was able to
navigate a maze (or corridors etc.). The robot, called Frank, uses a
number of specialised neural networks to process the information sent by
the sensors.
I am interested in the hardware of the robot. More specifically in the
way the neural networks exchange information and receive the signals from
the sensors. I am interested in the traffic of messages (?) between the
neural networks and a central controller, between the sensors and the
neural networks, and between neural networks, whatever is applicable.
I can't recall the name of the researcher or the institution where the
robot was developed. Can anyone point me to some reference or to the
authors.
Thanks in advance,
Paulo.
+-----------------------------+---------------------------------------------+
Paulo Valverde de L. P. Rocha | JANET:procha@uk.ac.ucl.cs
Department of Computer Science| BITNET:procha%uk.ac.ucl.cs@UKACRL
University College London |Internet:procha%cs.ucl.ac.uk@nsfnet-relay.ac.uk
Gower Street | ARPANet:procha@cs.ucl.ac.uk
London WC1E 6BT | UUCP:...!mcvax!ukc!ucl-cs!procha
England | tel: +44 (071) 387 7050 x 3719
| fax: +44 (071) 387 1397
+-----------------------------+---------------------------------------------+
------------------------------
Subject: Smalltalk Neural Network Program
From: SNYDERR@randd.abbott.com
Date: Thu, 21 Nov 91 08:51:00 -0600
Does anyone have or heard of a neural network program written in the
Smalltalk language? Thanks in advance.
Bob Snyder <snyderr@randb.abbott.com>
------------------------------
Subject: Frank, the robot (?). The gathered information.
From: Paulo V Rocha <P.Rocha@cs.ucl.ac.uk>
Date: Mon, 25 Nov 91 07:49:34 +0000
[[ Re: the BBC TV programme on neural networks and a robot called Frank. ]
- ------- Forwarded Messages
Subject: Re: A robot called Frank.
From: jon@uk.ac.oxford.robots (Jon Tombs)
To: P.Rocha@uk.ac.ucl.cs
Date: Mon, 25 Nov 91 02:01:30 +0000
The company is called TAG (Technology Application Group?), the managing
director is currently working for BT or some other big firm now, I don't
think they will be trading for much longer. Frank is a sort of neural
subsumption architecture, it has low level networks (analogue multiplier
implemented, "trained by hand"), to keep it in the middle of the path, and
higher levels to recognise T-junctions to turn right, dead-ends to turn
around etc, and with these it can solve a fixed maze (it doesn't solve them,
its logic is such that it always turns the way towards the centre).
PV >> Another information is that the robot seems to just follow a
PV >> pre-programmed map of the maze. Therefore, in my opinion, it doesn't
PV >> solve anything. It only navigates the map.
Hope this is of help, don't read to much into what you see on TV...
PV >> Well, I've taken vaccine against hype. :-)
- ------- Message 2
Subject: Frank
From: Mahesan Niranjan <niranjan@uk.ac.cambridge.engineering>
To: P.Rocha@uk.ac.ucl.cs
Date: Wed, 20 Nov 91 14:01:15 +0000
I saw that program too, don't remember where Frank is from. But are you
aware of the work at Oxford (Lionel Tarassenko) on neural net robot.
Their robot talks to a remote controller.
I think it was published in one of the NIPS (possibly the latest one)
Hope that helps
niranjan
- ------- End of Forwarded Messages
------------------------------
Subject: Postdoctoral work at Caltech
From: Jim Bower <jbower@cns.caltech.edu>
Date: Mon, 25 Nov 91 10:01:38 -0800
Postdoctoral Position in Computational Neurobiology
Computation and Neural Systems Program
Caltech
A Post-doctoral position in the laboratory of Dr. Gilles Laurent is
available for up to 3 years Applicants should have experience in
modelling techniques and be interested in general problems of
sensory-motor integration and/or single neuron computation. One possible
project would focus on somatosensory processing in insects emaphasizing
the architecture of local circuits comprised of a few hundred identified
neurons. These circuits are composed of 4 layers of neurons (sensory,
interneuronal 1, interneuronal 2, motor), with a large degree of
convergence and no known internal feedback connections. The task which
these circuits perform is the mediation of leg reflexes, and the
adaptation of these reflexes to external inputs or to internal
constraints (e.g. centrally generated rhythm). The second possible
project would focus on the integrative properties of the 2 classes of
local interneurons in those circuits. Both classes lack an axon (they
are local neurons), but the first ones use action potentials whereas the
second use graded potentials as modes of intra- and inter-cellular
communication. The hypothesis which we are trying to test experimentally
is that graded processing allows compartmentalization of function,
thereby increasing the computational capabilities of single neurons.
For further information contact:
Gilles Laurent
Biology Division
CNS Program, MS 139-74
Caltech
Pasadena, CA 91125
(818) 397-2798
laurent@delphi.caltech.edu.
------------------------------
Subject: Cognitive/Experimental Psychologist position available
From: "HQHSD::MRGATE::"A1::LAUER"" <lauer"%hqhsd.mrgate."a1.decnet@hsdp2.brooks.af.mil>
Date: 26 Nov 91 12:19:00 -0600
Senior Scientist/Cognitive Psychologist: Modern Technologies Corp., seeks
to fill the position in San Antonio, TX. Ph.D. in cognitive or
experimental is required. The successful candidate will have an interest
in cognitive task analysis and training program evaluation. MTC is a
high-technology company that offers a variety of technical, engineering,
and manufacturing services to government and industry. Salary for this
permanent position depends on qualifications. Send resume and salary
history to: Mr. Stan Foltz, MTC, 1313 SE Military Drive, Ste. 110, San
Antonio, TX 78214.
------------------------------
Subject: Cognitive Science at Birmingham
From: D.M.Peterson@computer-science.birmingham.ac.uk
Date: Fri, 29 Nov 91 15:10:43 +0000
============================================================================
University of Birmingham
Graduate Studies in COGNITIVE SCIENCE
============================================================================
The Cognitive Science Research Centre at the University of Birmingham
comprises staff from the Departments/Schools of Psychology, Computer
Science, Philosophy and English, and supports teaching and research in
the inter-disciplinary investigation of mind and cognition. The Centre
offers both MSc and PhD programmes.
MSc in Cognitive Science
The MSc programme is a 12 month conversion course, including a 4 month
supervised project. The course places a particular stress on the
relation between biological and computational architectures.
Compulsory courses: AI Programming, Overview of Cognitive
Science, Knowledge Representation Inference and Expert Systems, General
Linguistics, Human Information Processing, Structures for Data and
Knowledge, Philosophical Questions in Cognitive Science, Human-Computer
Interaction, Biological and Computational Architectures, The Computer and
the Mind, Current Issues in Cognitive Science.
Option courses: Artificial and Natural Perceptual Systems,
Speech and Natural Language, Parallel Distributed Processing.
It is expected that students will have a good first degree ---
psychology, computing, philosophy or linguistics being especially
relevant.
Funding is available through SERC and HTNT.
PhD in Cognitive Science
For 1992 studentships are expected for PhD level research into a range
of topics including:
o computational modelling of emotion
o computational modelling of cognition
o interface design
o computational and psychophysical approaches to vision
Computing Facilities
Students have access to ample computing facilities, including networks
of Hewlett-Packard, Sun and Sparc workstations in the Schools of Computer
Science and Psychology.
Contact
For further details, contact: The Admissions Tutor, Cognitive Science,
School of Psychology, University of Birmingham, PO Box 363, Edgbaston,
Birmingham B15 2TT, UK.
Phone: (021) 414 3683
Email: cogsci@bham.ac.uk
------------------------------
Subject: connectionism and social psychology
From: falex@loria.crin.fr (Frederic Alexandre)
Organization: CRIN (CNRS) Nancy - INRIA Lorraine
Date: 30 Nov 91 10:42:20 +0000
[[ Editor's Note: Hmmm, I'm also curious about the "connection" between
these two disparate disciplines! -PM ]]
Hi, has anyone heard of papers or works that try to relate connectionism
and social psychology? (e.g lateral inhibition or competition considered
as group behavior) Thank you for your help
alex
E-mail: falex@loria.crin.fr
------------------------------
Subject: reminder and solicitation of INNS vote
From: Andras Pellionisz SL <pellioni@pioneer.arc.nasa.gov>
Date: Mon, 02 Dec 91 11:31:29 -0800
[[ Editor's Note: It is with some trepidation that I include this entry.
While I respect Pellionisz' intentions, I am uncomfortable about allowing
this scientific informational forum to be used for personal campaigning.
In the past, I have rejected commercial announcements and submissions of
resumes (curriculum vitae) for people looking for positions or selling
something. However, since last year, I have been taking a more liberal
stance for editing. I would appreciate feedback from readers about the
appropriateness of this notice. Should I have rejected it or are you glad
that I included it? -PM ]]
*******************************************
******** SEND IN YOUR INNS VOTE TODAY *********
******A REMINDER AND VOTE-SOLICITATION********
*******************************************
**************BY A.J. PELLIONISZ **************
*******************************************
Dear Fellow Member in the International Neural Network Society!
May I remind you that the 1991 Ballot for Officers of the INNS has to
be postmarked by December 12th ?
I ALSO ASK YOU TO VOTE FOR ME FOR THE BOARD OF DIRECTORS.
With the dawn of a $15 million "Neurocomputing Project" (fiscal YR
93-98), NASA is very close to becoming a central force in the US Civilian
Neurocomputing Initiative. What is more, aerospace applications (and
Aerospace Agencies in the US and Worldwide) emerge as a forum that could
make neurocomputers the Aerospace Computers of the Future (where
error-tolerance, fast parallel computation and adaptive software with
massively reduced complexity are vital). Aerospace goals are paramount.
National and regional aerospace agencies (such as NASA in the USA) could
well become the system to manage the internationally competitive and
cooperative aspects of neurocomputing.
I need your help. On the Board of Directors of International Neural
Network Society, I will work to make sure that neurocomputer research and
development fully benefits from the above strategy.
>From a basic research point of view, many of you may have followed
recent developments of my "geometrical approach" to neurocomput- ing. Now
it is internationally recognized. In 1990 it was awarded by the Alexander
von Humboldt Prize by Germany. In 1991, in Japan, an early critique of
the approach, S. Amari, have turned around and at this time champions a
tensor-geometrical approach which "opens a new fertile field in NN
research" (Neural Networks, 1991 Sept). In fact, I will place an "open
letter to S. Amari" in a forthcoming issue of "Neuron Digest" on some
aspects of pioneering this approach in the past decade.
Many of you know that I always put great emphasis on the neuro-
science-roots of the basic-research aspects of neurocomputing. Now you
wittness that we learn mathematical (geometrical) conclusions from the
study of existing neural nets. At this time, application of
neurocomputing, in a field of computer science and engineering (in
aerospace instrumentation) unfolds, opening great potential for all of
us.
Please come forward with your vote to help me coordinate such activities
by means of the INNS! It was conceived for this role!
Thanking for your vote,
Sincerely
Andras J. Pellionisz
Professor of Biophysics at New York University
Senior National Research Council Associte to NASA by
National Academy of Sciences
NASA Ames Res. Ctr., Neurocomputing Lab., 261-3
Moffett Field, CA 94035
Phone: 415-604-4821
Fax: 415-604-0046
E-mail: Pellioni@Pioneer.arc.nasa.gov
If you misplaced your ballot-sheet, please find below a "reminder":
PRESIDENT
Dr. Paul Werbos
First Choice (check 1) Second Choice (check 1)
PRESIDENT-ELECT
Dr. Judith E. Dayhoff
Dr. Bart Kosko
Dr. H. Harold Szu
GOVERNING BOARD (vote for six)
Dr. Andrew Barto
Dr. Judith Dayhoff
Dr. Walter Freeman
Dr. Dan Hammerstrom
Prof. Morris Hirsch
Dr. James C. Houk
Dr. Mitsuo Kawato
Prof. Teuvo Kohonen
Dr. Bart Kosko
V.R. Narendra
Robert Pap
Prof. Andras J. Pellionisz
Ballot-sheet (or the above equivalent) must be returned to
INNS
1250, 24th Street, NW
Suite 300
Washington DC 20037
USA
Validate your ballot by putting your NAME, SIGNATURE and
MEMBERSHIP # (if at hand) on the ENVELOPE (for confidentiality).
Ballots must be postmarked by December 12, 1991
SEND IN YOUR VOTE TODAY !
If you have any questions, call INNS at (202) INN-INNS.
Thanking your vote of confidence in me in advance!
------------------------------
Subject: job ad for NIPS message board
From: Peter Killeen <ICPRK%ASUACAD.BITNET@BITNET.CC.CMU.EDU>
Date: Mon, 02 Dec 91 14:40:59 -0700
Experimental Psychologist
Arizona State University is recruiting an Associate Professor of
Experimental Psychology. The successful candidate must have a Ph.D. in
psychology with specialty in adaptive systems/neural networks. The
individual will join a growing interdisciplinary group with interests in
basic biomedical research from an adaptive systems perspective. The
position will start in August 1992. Send Vitae, reprints, and contact
three individuals to have them send letters of reference to Dr. Peter R.
Killeen, Experimental Psychology Search Committee, Department of
Psychology, Arizona State University, Tempe, AZ 85287-1104. Deadline for
application is Feb 14, 1992, and every two weeks thereafter until the
position is filled. ASU is an equal- opportunity and affirmative action
employer.
------------------------------
Subject: Graduate and Post-doc positions in Neural Processes in Cognition
From: SCHNEIDER@vms.cis.pitt.edu
Date: Tue, 03 Dec 91 12:23:00 -0500
Program announcement for
Interdisciplinary Graduate and Postdoctoral Training
in Neural Processes in Cognition
at the University of Pittsburgh and Carnegie Mellon University
The Pittsburgh Neural Processes in Cognition program, in its second year
is providing interdisciplinary training in brain sciences. The National
Science Foundation has established an innovative program for students
investigating the neurobiology of cognition. The program's focus is the
interpretation of cognitive functions in terms of neuroanatomical and
neurophysiological data and computer simulations. Such functions include
perceiving, attending, learning, planning, and remembering in humans and
in animals. A carefully designed program of study prepares each student
to perform original research investigating cortical function at multiple
levels of analysis. State of the art facilities include: computerized
microscopy, human and animal electrophysiological instrumentation,
behavioral assessment laboratories, MRI and PET brain scanners, the
Pittsburgh Supercomputing Center, and a regional medical center providing
access to human clinical populations. This is a joint program between
the University of Pittsburgh, its School of Medicine, and Carnegie Mellon
University.
Each student receives full financial support, travel allowances and a
computer workstation.
Applications are encouraged from students with interest in biology,
psychology, engineering, physics, mathematics, or computer science. Last
year's class included mathematicians, psychologists, and neuroscience
researchers.
Pittsburgh is one of America's most exciting and affordable cities,
offering outstanding symphony, theater, professional sports, and outdoor
recreation in the surrounding Allegheny mountains. More than ten
thousand graduate students attend its universities.
Core Faculty and interests and affiliation
Carnegie Mellon University -Psychology- James McClelland, Johnathan
Cohen, Martha Farah, Mark Johnson
University of Pittsburgh
Behavioral Neuroscience - Michael Ariel
Biology - Teresa Chay
Information Science - Paul Munro
Mathematics - Bard Ermentrout
Neurobiology Anatomy and Cell Sciences - Al Humphrey
Neurological Surgery - Don Krieger, Robert Sclabassi
Neurology - Steven Small
Psychiatry - David Lewis, Lisa Morrow, Stuart Steinhauer
Psychology - Walter Schneider, Velma Dobson
Physiology - Dan Simons
Radiology - Mark Mintun
Applications:
To apply to the program contact the program office or one of the
affiliated departments. Students are admitted jointly to a home
department and the Neural Processes in Cognition Program. Postdoctoral
applicants must have United States resident's status. Applications are
requested by February 1. For information contact:
Professor Walter Schneider
Program Director
Neural Processes in Cognition
University of Pittsburgh
3939 O'Hara St
Pittsburgh, PA 15260
Or: call 412-624-7064 or Email to
NEUROCOG@VMS.CIS.PITT.BITNET.
------------------------------
Subject: Invariant pattern recognition with ANNs
From: evol@infko.uni-koblenz.de
Organization: University of Koblenz, Germany
Date: 05 Dec 91 14:34:33 +0000
I am interested in translation, rotation and scaleing invariant pattern
recognition with ANNs. I already know about Fokushima's Neocognitron and
Hubel & Wiesels work about the human visual cortex. I also read some
papers about complex logarithmic mapping and Fast Fourier Transformation
(H. Haken, H. J. Reitboeck & J. Altmann, H. Wechsler & G. L. Zimmerman).
Nevertheless this complex logarithmic mapping seems to take place in the
human visual cortex, I think it is not sufficient for the task.
Especially scaleing invariant recognition is still a very hard problem.
Are there any additional related works about scaleing invariant pattern
recognition ?
Thanks in advance
Randolf Werner
Randolf Werner FB Informatik, Uni Koblenz, evol@infko.UUCP
Rheinau 3-4, D-5400 Koblenz, Germany ...!uunet!mcsun!unido!infko!evol
------------------------------
End of Neuron Digest [Volume 8 Issue 13]
****************************************