Copy Link
Add to Bookmark
Report
Neuron Digest Volume 12 Number 17
Neuron Digest Thursday, 18 Nov 1993 Volume 12 : Issue 17
Today's Topics:
IWANNT-EPROCS - note that the correct public login is iwan_pub
Need refs on Spherical Harmonics
Oxford Connectionist Summer School
AI CD-ROM Revision 2 - press release
Adaptive Simulated Annealing (ASA) version 1.53
Neuron Submission
request for information
Academic position at USC
Re: Whence cybernetics
Vclamp and Cclamp programs?
Looking for a paper from some conference
Send submissions, questions, address maintenance, and requests for old
issues to "neuron-request@psych.upenn.edu". The ftp archives are
available from psych.upenn.edu (130.91.68.31). Back issues requested by
mail will eventually be sent, but may take a while.
----------------------------------------------------------------------
Subject: IWANNT-EPROCS - note that the correct public login is iwan_pub
From: Bob Allen <rba@bellcore.com>
Date: Tue, 09 Nov 93 12:45:40 -0500
Subject: IWANNT'93 Electronic Proceedings
Electronic Proceedings for
1993 International Workshop on Applications of Neural Networks
to Telecommunications
1. Electronic Proceedings (EPROCS)
The Proceedings for the 1993 International Workshop on
Applications of Neural Networks to Telecommunications
(IWANNT'93) have been converted to electronic form and are
available in the SuperBook(TM) document browsing system. In
addition to the IWANNT'93 proceedings, you will be able to
access abstracts from the 1992 Bellcore Workshop on
Applications of Neural Networks to Telecommunications and
pictures of several of the conference attendees.
We would appreciate your feedback about the use of this
system. In addition, if you have questions, or would like a
personal account, please contact Robert B. Allen
(iwannt_allen@bellcore.com or rba@bellcore.com).
2. Accounts and Passwords
Public access is available with the account name: iwan_pub
Individual accounts and passwords were given to conference
participants. Annotations made by iwan_pub may be edited by the
electonic proceedings editor.
3. Remote Access Via Xwindows
>From an Xwindow on machine connected to the Internet do the
following. Note that some locations may have "firewall"
that prevents Xwindows' applications from running. If this
procedure fails, you may have to find a machine outside your
firewall or use the character-based interface (csb).
+ xhost +128.96.58.4 (Xwindows display permission for
superbook.bellcore.com)
+ telnet 128.96.58.4
+ (login)
+ TERM=xterms (it is important to use "xterms")
+ enter your email address
+ Figure out and enter your machine's IP address
(in /etc/hosts or ask an administrator)
+ gxsb (Xwindows version of SuperBook)
3.1 Overview of Xwindows SuperBook Commands
When you login to SuperBook, you will obtain a Library
Window. For the IWANNT proceedings, you should select the
IWANNT shelf, highlight "Applications of Neural Networks to
Telecommunications" and click "Open". The Text Window
should be placed on the right side of the screen and the
Table-of-Contents Window should be placed on the left.
These windows can be resized.
Table-of-Contents (TOC): Books, articles, and sections
within books can be selected by clicking in the TOC. If the
entry contains subsections, it will be marked with a "+".
Double-clicking on those entries expands them. Clicking on
an expanded entry closes it.
Text Window: The text can be scrolled one-line-at-a-time
with the Scroll Bar Arrows or a page-at-a-time by clicking
on the spaces immediately above or below the Slider.
Graphics: Figures, tables, and some equations are presented
as bitmaps. The graphics can be viewed by clicking on the
blue icons at the right side of the text which pops up a
bitmap-viewer. Graphics can be closed by clicking on their
"close" button. Some multilemdia applications have been
included, but these may not work correctly across the
Internet.
Searching: Terms can be searched in the text by typing them
into the Search-Window. Wild-card searches are possible as
term* You can also search by clicking on a term in the text
(to clear that search and select another, do ctrl-click).
Annotations: Annotations are indicated with a pencil icon
and can be read by clicking on the icon. Annotations can be
created (with conference-attendee logins) by clicking in the
text with the left button and then typing in the annotation
window.
Exiting: Pull down the FILE menu on the Library Window to
"QUIT", and release.
4. Remote Access via character-based interface
>From any machine connected to the Internet do the following:
+ telnet 128.96.58.4 (for superbook.bellcore.com)
+ (login)
+ TERM=(termtype) (use "xterms" for an Xwindow inside
a firewall)
+ enter your email address
+ csb
4.1 Overview of csb SuperBook Commands
The character-based interface resembles emacs. You first
enter Library mode. After selecting a shelf (make sure you
are on the IWANNT shelf) and a book on that shelf (e.g.,
Applications of Neural Networks to Telecommunications), the
screen is split laterally into two parts. The upper window
is the TOC and the lower window has the text.
Table-of-Contents (TOC): Books, articles, and sections
within books can be selected by typing the number beside
them in the TOC. If the entry contains subsections, it will
be marked with a "+".
Text Window: The text can be scrolled one-line-at-a-time
with the u/d keys or a page-at-a-time with the U/D keys.
Graphics: Most bitmapped graphics will not be available.
Searching: Terms can be searched in the text by typing them
into the Search-Window. Wild-card searches are possible as
term* Searches are also possible by posting the cursor over
a word and hitting RET.
Annotations: Annotations are indicated with an A on the
right edge of the screen. These can be read by entering an
A on the line on which they are presented. Annotations can
be created (given correct permissions) by entering A on any
line.
Exiting: Enter "Q"
------------------------------
Subject: Need refs on Spherical Harmonics
From: slehar@copley.bu.edu (Steve Lehar)
Organization: Boston University Center for Adaptive Systems
Date: 10 Nov 93 18:10:59 +0000
RADICAL NEW BRAIN THEORY
I have come up with a very interesting new theory on the way the brain
represents spatial patterns by way of harmonic resonant interactions
between electrically coupled neurons. The big advantage of using such
resonances is that a simple physical system, for example a bell, is
capable of responding to and reproducing fantastically complex spatial
patterns, i.e. the harmonics of the bell, in a remarkably robust
manner using simple dynamical interactions, and can encode those
complex spatial patterns in a simple and highly compressed rotation
invariant code, i.e. the oscillation frequency corresponding to the
harmonic pattern. This is a radically new theory, and suggests an
entirely new mode of communication between neurons in the brain. I
have performed a number of computer simulations which have confirmed
the validity of the theory by accurately reproducing a large number of
visual phenomena by way of a single simple mechanism.
NEED REFERENCES ON SPHERICAL HARMONICS
I would now like to extend the orientational harmonic theory to
explain three-dimensional harmonics, in order to explore the kinds of
spatial structures that can be represented by this kind of system. In
order to do this, I need to read up on spherical harmonics. I have
found a number of books in the library full of fantastically elaborate
differential equations, etc, but that is NOT what I am looking for.
What I want is either a catalogue of the kinds of patterns represented
by the first few harmonics, or a SIMPLE description of how I could
write a computer program that would reproduce these patterns.
For example, some astronomy books explain how the sun is constantly
oscillating, or ringing like a bell, and the fundamental harmonics of
these oscillations include:
bulging all over, then shrinking all over, alternately
bulging in the northern hemisphere while shrinking in the
southern, and the inverse, alternately
bulging at the equator and shrinking at the poles...
bulging at the equatior and at a northern and southern "temperate"
lattitude, while shrinking elsewhere...
etc, etc...
in other words, this set of harmonics defines a series of
alternating subdivisions by lattitude. Another set of harmonics
defines a similar alternation by longitude, creating alternating
"orange slices". Combinations of these two modes of oscillation
produce checkerboard patterns, and so forth. Each of these patterns
corresponds to a particular temporal waveform or oscillation
frequency which therefore represents that pattern in a rotation
invariant manner. The bizzar patterns seen in electron orbitals are
yet another example of spherical harmonics. Does anybody know of a
reference that would list these different pattern types, or give
equations that are simple enough that I could type them into my
computer and generate families of these patterns myself? I have
neither the inclination nor the ability to plummet the depths of the
differential equations defining such systems, I only want to look at
the resultant patterns.
Can anybody out there help me?
- --
(O)((O))(((O)))((((O))))(((((O)))))(((((O)))))((((O))))(((O)))((O))(O)
(O)((O))((( slehar@park.bu.edu )))((O))(O)
(O)((O))((( Steve Lehar Boston University Boston MA )))((O))(O)
(O)((O))((( (617) 424-7035 (H) (617) 353-6741 (W) )))((O))(O)
(O)((O))(((O)))((((O))))(((((O)))))(((((O)))))((((O))))(((O)))((O))(O)
------------------------------
Subject: Oxford Connectionist Summer School
From: plunkett (Kim Plunkett) <@prg.ox.ac.uk:plunkett@dragon.psych>
Date: Thu, 11 Nov 93 18:43:43 +0000
UNIVERSITY OF OXFORD
MRC BRAIN AND BEHAVIOUR CENTRE
McDONNELL-PEW CENTRE FOR COGNITIVE NEUROSCIENCE
SUMMER SCHOOL ON CONNECTIONIST MODELLING
Department of Experimental Psychology
University of Oxford
11-23 September 1994
Applications are invited for participation in a 2-week
residential Summer School on techniques in connectionist
modelling of cognitive and biological phenomena. The course
is aimed primarily at researchers who wish to exploit neural
network models in their teaching and/or research. It will
provide a general introduction to connectionist modelling
through lectures and exercises on PCs. The instructors with
primary responsibility for teaching the course are Kim
Plunkett and Edmund Rolls.
No prior knowledge of computational modelling will be
required though simple word processing skills will be
assumed. Participants will be encouraged to start work on
their own modelling projects during the Summer School.
The Summer School is sponsored (jointly) by the University
of Oxford McDonnell-Pew Centre for Cognitive Neuroscience
and the MRC Brain and Behaviour Centre. The cost of parti-
cipation in the summer school is 500 pounds to include
accommodation (bed and breakfast at St. John's College) and
summer school registration. Participants will be expected to
cover their own travel and meal costs. A small number of
graduate student scholarships may be available. Applicants
should indicate whether they wish to be considered for a
graduate student scholarship but are advised to seek their
own funding as well, since in previous years the number of
graduate student applications has far exceeded the number of
scholarships available.
If you are interested in participating in the Summer School,
please contact:
Mrs. Sue King
Department of Experimental Psychology
University of Oxford
South Parks Road
Oxford OX1 3UD
Tel: (0865) 271353
Email: sking@uk.ac.oxford.psy
Please send a brief description of your background with an
explanation why you would like to attend the Summer School
(one page maximum) no later than 1 April 1994.
------------------------------
Subject: AI CD-ROM Revision 2 - press release
From: ncc@ncc.jvnc.net (R. Steven Rainwater)
Date: Sat, 13 Nov 93 17:48:32 -0600
[[ Editor's Note: I have edited this message heavily -- in fact, deleted
the press release. It is close to my threshold for a "commercial
announcement"(even if *does* contain the archives of Neuron Digest), so I
will let interested readers get the table of contents files and/or
contact the poster. The official retail cost for this CD-ROM is $129,
though it seems that most of the contents are generally distributable
without license. -PM ]]
We actually started shipping the Rev.2 discs a couple of months ago but due
to the number of orders already waiting to be filled we didn't want to send
out press releases until we'd caught up. Anyway, things seem to be going
along fine at this point, so here's the press release! The hard copies are
going out to the trade mags at about the same time, so they'll probably
start showing up in a month or so.
For people wanting to see a complete listing of the CD's contents, look for
the file AICDROM2.ZIP at an ftp site near you. The file is also available
from the Compuserve AI forum, and the NCC dial-up BBS at 214-258-1832. It
contains the file listing, this press release, a couple of magazine reviews
of the disc, and other assorted information. Oh, and one last note regarding
pricing - customers have told us that Programmer's Pardise seems to have the
best pricing in the US on the disc...
- -Steve Rainwater
============================================================================
Network Cybernetics Corporation -- Press release for electronic distribution
============================================================================
ANNOUNCING THE AI CD-ROM REVISION 2
Network Cybernetics Corporation is now shipping the second annual revision
of their popular AI CD-ROM, an ISO-9660 format CD-ROM containing a wide
assortment of information on AI, Robotics, and other advanced machine
technologies. The AI CD-ROM contains thousands of programs, source code
collections, tutorials, research papers, Internet journals, and other
resources. The topics covered include artificial intelligence, artificial
life, robotics, virtual reality, and many related fields. Programs for OS/2,
DOS, Macintosh, UNIX, Amiga, and other platforms can be found on the disc.
The files have been collected from civilian and government research centers,
universities, Internet archive sites, BBS systems and other sources. The
CD-ROM is updated annually to keep it current with the latest trends and
developments in advanced machine technologies such as AI.
[[ the rest of this text deleted for brevity. -PM ]]
------------------------------
Subject: Adaptive Simulated Annealing (ASA) version 1.53
From: Lester Ingber <ingber@alumni.cco.caltech.edu>
Date: Sun, 14 Nov 93 03:15:46 -0800
========================================================================
Adaptive Simulated Annealing (ASA) version 1.53
To get on or off the ASA email list, just send an email to
asa-request@alumni.caltech.edu with your request.
________________________________________________________________________
Significant CHANGES since 1.43 (17 Sep 93)
In General Information below, note the changes in the names of the
files containing the ASA code in the archive. The original filename
was decided by Netlib; with the code now in the Caltech archive, more
standard conventions can be used.
In asa.c, INT_ALLOC and INT_LONG defines had to be corrected around a
few fprintf statements. This may affect some PCs.
On a Sun, gcc-2.5.0 update runs were performed. Since the change in
definitions of MIN_DOUBLE and MAX_DOUBLE, setting SMALL_FLOAT=E-12 does
not agree with SMALL_FLOAT=E-18 and SMALL_FLOAT=E-20 (the latter two
agree), unless MIN_DOUBLE=E-18 also is set. The results diverge when
the parameter temperatures get down to the range of E-12, limiting the
accuracy of the SMALL_FLOAT=1.0E-12 run.
Added TIME_STD Pre-Compile Option to use unix-standard macros for time
routines, as required by some systems, e.g., hpux.
The ASA_TEMPLATE and ASA_TEST OPTIONS were added to permit easier use
of templates and test examples. Setting ASA_TEST to TRUE will permit
running the ASA test problem. Searching user.c for ASA_TEST also
provides a guide to the user for additional code to add for his/her own
system. Keeping the default of ASA_TEST set to FALSE permits such
changes without overwriting the test example. There are several
templates that come with the ASA code, used to test several OPTIONS.
To permit use of these OPTIONS without having to delete these extra
tests, these templates are wrapped with ASA_TEMPLATE.
________________________________________________________________________
ASA-Related Papers
The following two papers have used ASA to solve some very difficult
imaging problems that did not yield to other global optimization
techniques.
%A G. Blais
%A M.D. Levine
%T Registering multiview range data to create 3D computer objects
%R TR-CIM-93-16
%I Center for Intelligent Machines, McGill University
%C Montreal, Canada
%D 1993
This paper is a large file, 1.7 MBytes gzip'd, and there may soon be an
ftp site available for its retrieval. You can contact Gerard Blais
<gblais@mcrcim.mcgill.edu> for further information.
%A K. Wu
%A M.D. Levine
%T 3-D object representation using parametric geons
%R TR-CIM-93-13
%I Center for Intelligent Machines, McGill University
%C Montreal, Canada
%D 1993
This paper can be retrieved via anonymous ftp as
object_representation.ps.gz from ftp.caltech.edu in directory
pub/ingber/.limbo. (Read file INDEX in pub/ingber to retrieve files
from .limbo.) Contact Wu Kenong <wu@mcrcim.mcgill.edu> for further
information.
I believe that files should be stored in archives related more to a
relevant discipline than to the computational tools used to derive
results. So, while I can document ASA-related papers and actively
search for ftp sites to hold electronic (p)reprints, I cannot guarantee
storage for them in the Caltech ASA archive. The problem is that often
it is difficult to find archives for large files, e.g., containing
scanned figures. If anyone has authority to house large files to be
made available via anonymous ftp, please let me know what disciplines
you are willing to support.
________________________________________________________________________
Wall Street Journal
The reference to the article in the WSJ that mentioned the wide-spread
use of the ASA code is M. Wofsey, "Technology: Shortcut Tests Validity
of Complicated Formulas," The Wall Street Journal, vol. CCXXII, no. 60,
p. B1, 24 September 1993.
As I stated in the last general update: I gave the WSJ examples of
some projects using ASA, but I had to insist that the relevant people
would have to be contacted previous to citing them. Of course the
press has the last word on what they will publish/interpret.
________________________________________________________________________
General Information
The latest Adaptive Simulated Annealing (ASA) code and some related
(p)reprints in compressed PostScript format can be retrieved via
anonymous ftp from ftp.caltech.edu [131.215.48.151] in the pub/ingber
directory.
Interactively: ftp ftp.caltech.edu, [Name:] anonymous, [Password:]
your_email_address, cd pub/ingber, binary, ls or dir, get
file_of_interest, quit. The INDEX file contains an index of the other
files and information on getting gzip and unshar for UNIX, DOS and MAC
systems.
The latest version of ASA is ASA-x.y-shar.Z (x and y are version
numbers), linked to ASA-shar.Z. For the convenience of users who do
not have any uncompress/gunzip utility, there is a file ASA-shar which
is an uncompressed copy of ASA-x.y-shar.Z/ASA-shar.Z; if you do not
have sh or shar, you still can delete the first-column X's and separate
the files at the END_OF_FILE locations. For the convenience of some
users, there also is a current gzip'd tar'd version, ASA-x.y.tar.gz,
linked to ASA.tar.gz. There are patches ASA-diff-x1.y1-x2.y2.Z up to
the present version; these may be concatenated as required before
applying. Only current patches may be available. If you require a
specific patch that is not contained in the archive, contact
ingber@alumni.caltech.edu.
If you do not have ftp access, get information on the FTPmail service
by: mail ftpmail@decwrl.dec.com, and send only the word "help" in the
body of the message.
If any of the above are not possible, and if your mailer can handle
large files (please test this first), the code or papers you require
can be sent as uuencoded compressed files via electronic mail. If you
have gzip, resulting in smaller files, please state this.
Sorry, I cannot assume the task of mailing out hardcopies of code or
papers. My volunteer time assisting people with their their queries on
my codes and papers must be limited to electronic mail correspondence.
Lester
========================================================================
|| Prof. Lester Ingber 1-800-L-INGBER ||
|| Lester Ingber Research Fax: [10ATT]0-700-L-INGBER ||
|| P.O. Box 857 EMail: ingber@alumni.caltech.edu ||
|| McLean, VA 22101 Archive: ftp.caltech.edu:/pub/ingber ||
------------------------------
Subject: Neuron Submission
From: "K. Mansell" <csp35@teach.cs.keele.ac.uk>
Date: Tue, 16 Nov 93 18:32:48 +0000
[[ Editor's Note: As readers know, I publish "jobs offered" but not "jobs
wanted". I've asked this fellow to rewrite his request, since it seemed
like an academic "internship." In the U.S., such positions are usually
volunteer, unpaid, and for a limited duration. Well, here it is... -PM ]]
I am a physics graduate in the UK, currently studying for an MSc in Machine
Perception and Neurocomputing. The second half of my course involves a 6
month project and I am in the process of contacting companies in the UK who
are working in the area of neural computing to see if they would be
agreeable to me working on a project for them.
The level of commitment/support provided by the company is at the discretion
of the company itself. The placement I am looking for would preferably be
at the company's site although it need not be paid work. I am interested
in the application of neural nets to real scientific/technical problems
and would rather not work on defence projects.
I would be grateful for any information (contacts, areas of work etc)
relating to companies working with neural networks in the UK who
might consider offering me a project placement. In particular, I am trying
to locate a company called Neural Computer Systems (Sciences?) who have so
far eluded my searching.
Kevin Mansell
<csp35@teach.cs.keele.ac.uk>
------------------------------
Subject: request for information
From: moody_k@orgella.com
Date: Tue, 16 Nov 93 16:50:24 -0500
Greetings!
I have just joined the neural digest mailing list, and I have
been excited to see so much activity. I am a graduate student
at the University of New Hampshire in electrical engineering,
and a relative newcomer to neural network technology.
I have been researching neural networks as related to control
systems. and have found many references to work done by
Narendra and Parthasarathy which was published in IEEE
transactions in March 1990, where they proposed combining
recurrent and multilayer networks, and training with
dynamic backpropagation.
I am wondering if anyone could recommend more recent references
which might indicate what these proposals have lead to, and
what the present state of these techniquexs may be.
Thank you in advance.
Kris Moody
moody_k@orgella.uucp
------------------------------
Subject: Academic position at USC
From: Jean-Marc Fellous <fellous@rana.usc.edu>
Date: Tue, 16 Nov 93 14:24:57 -0800
Could you please post this announcement ....
ASSISTANT/ASSOCIATE PROFESSOR
BIOMEDICAL ENGINEERING/NEUROSCIENCE
UNIVERSITY OF SOUTHERN CALIFORNIA
A tenure-trace faculty position is available in the Department
of Biomedical Engineering at the University of Southern California.
This is a new position, created to strengthen the concentration of
neuroscience research within the Department. Applicants should be
capable of establishing an externally funded research program that
includes a rigorous, quantitative approach to functional aspects of
the nervous system. A combined theoretical and experimental approach
is preferred, though applicants withpurely theoretical research
programs will be considered. Multiple opportunities for
interdisciplinary research are fostered by USC academic and research
programs such as the Biomedical Simulations Resource, the Program in
Neu~science, and the Center for Neural Computing. Send curriculum
vitae, three letters of recommendation, and a description of current
and future research by January 1, 1994 to
Search Committee,
Department of Biomedical Engineerig,
530 Olin Hall,
University of Southern California,
Los Angeles, CA 90089-1451.
------------------------------
Subject: Re: Whence cybernetics
From: Jim Brakefield <braker@ennex1.eng.utsa.edu>
Date: Wed, 17 Nov 93 09:46:49 -0800
The Sept. 93 issue of Connections mentions the Neuron Digest discussion on the AI/Cybernetic
history. I would like to comment.
I did a poster paper for IJCNN 1988 in which one topic was the development of quantitative "scales"
from qualitative concepts (i.e like the emmergence of a temperature scale from the three states of
matter). One example in the poster paper dealt with the emergence of two scales:
The subject area is AI/connectionism/expert systems and the like. The two scales are computation
requirements and memory requirements. The historical perspective is that early AI attempted to
minimize both computation and memory. The result was theorem proving (efficiency of the symbolic
and logic approach in both knowledge and calculation via derivation of formulas). The next epoch
was Expert Systems which allowed a larger knowledge data base (greater consumption of memory).
The current epoch of connectionism allows both large knowledge data and extensive computation.
Thus the progression of "AI" paradgms has followed the historical improvement in computer
ecconomy. The perspective is that of placing each AI algorithm at a position in a two dimensional
space of computational requirements and memory requirements. The early failure of the "cybernetic"
approach is thus a matter of its implementation requirements being ahead of its time. One can wonder
if Minsky and Papert book was an emergent phenonena resulting from the computer ecconomics
of its day.
James C. Brakefield braker@ennex1.eng.utsa.edu
------------------------------
Subject: Vclamp and Cclamp programs?
From: ilya@cheme.seas.upenn.edu (Ilya Rybak)
Date: Wed, 17 Nov 93 20:55:10 -0500
Dear friends,
I would like to get Vclamp and Cclamp programs for simulation of
single neuron that were developed and are distributing by
John Huguenard and David McCormick. I will be very thankful to
somebody for information (e-mail address) about how to order these
programs.
Ilya Rybak
ilya@cheme.seas.upenn.edu
------------------------------
Subject: Looking for a paper from some conference
From: "r.kelly" <DFCA4601G@UNIVERSITY-CENTRAL-ENGLAND.AC.UK>
Date: Thu, 18 Nov 93 17:06:38
Hello
I have been looking (unsuccessfully) for the following paper for some
time now :
D Nguyen, B Widrow, "Improving the learning speed of 2-layer neural
networks by choosing initial values of the adaptive weights",
International Conference of Neural Networks, July 1990
I don't have access to any Neural Network conference proceedings and
the ILL counter at my library is unable to track it down. Can anyone
tell me which International Conference is being referred to here?
Thanks,
Heather
Internet : dfca4601g%uk.ac.uce@nsf.ac.uk
Janet : dfca4601g@uk.ac.uce
------------------------------
End of Neuron Digest [Volume 12 Issue 17]
*****************************************
