Copy Link
Add to Bookmark
Report
Neuron Digest Volume 08 Number 02
Neuron Digest Tuesday, 8 Oct 1991 Volume 8 : Issue 2a
Today's Topics:
Administrivia - Miss-numbered issue and ftp archives
positions
we want your messages!
NN s/w for 80x86 machines?
New Journal and CFP
Re: Welcome to Neuron-Digest
Request for meeting in Hawaii
Request for info on neural nets and forecasting
Re: Neuron Digest V8 #1
IJCNN '91 Singapore - Request to share a room
Re: Review of Handwritten Character Recognition
Brain Design
walking robots
tree classification code available for comparative studies
Re: Neuron Digest V8 #1
Send submissions, questions, address maintenance and requests for old
issues to "neuron-request@cattell.psych.upenn.edu". The ftp archives are
not yet available. Back issues requested by mail will eventually be sent,
but may take a while. Full service should resume shortly. Stay tuned.
----------------------------------------------------------------------
Subject: Administrivia - Miss-numbered issue and ftp archives
From: "Neuron-Digest Moderator, Peter Marvit" <neuron@cattell.psych.upenn.edu>
Date: Tue, 08 Oct 91 15:08:21 -0500
As many of you astutely observed, the last issue said Vol 8 #1 in the
Subject: line, but #2 in the body of the Digest. It really is #1 and has
been corrected in archives. I will label this issue #2a, but it really is
just #2.
Requests for mailed back issues have not been fulfilled and probably will
be "on hold" for a while still. Sorry. I hope to have an automated
archive server for Neuron Digest ready in a month or so (between my
regular research, my papers, and still unpacking).
However, the ftp archives should now be working. You can ftp old issues
of Neuron Digest, as well as a small sampling of software from
cattell.psych.upenn.edu (128.91.2.173). Look in pub/Neuron-Digest for
back issues and pub/Neuron-Sowftare for code. One of these days, I will
try to put pointers to other sources for code in that directory.
Contributions are always welcome. Contact me for more information.
UUCP sites may still be experiencing difficulties getting Neuron Digest.
If you were a UUCP address and can be reached via Internet, *please* send
me a change of address.
Dendtrically yours,
Peter Marvit
: Peter Marvit, Neuron Digest Moderator
: Courtesy of the Psychology Department, University of Pennsylvania
: neuron-request@cattell.psych.upenn.edu
------------------------------
Subject: positions
From: Susan Newman <snewman@parc.xerox.com>,
mcole@weber.ucsd.edu (Mike Cole)
Date: Sun, 11 Aug 91 12:14:45 -0800
[[ Editor's Note. Please contact the address noted in the message. Do
NOT send queries to the orginal poster. -PM ]]
Please circulate and forward this advertisement far and wide.
don norman
=------------------------------------------------------------------------
TWO POSITIONS IN COGNITIVE SCIENCE
COGNITIVE SCIENCE RESEARCH IN
MATH & SCIENCE EDUCATION
The Department of Cognitive Science at the University of California, San
Diego seeks someone in the area of learning in mathematics and the
sciences. Candidates must have demonstrated interest in cognitive
science and education. The position is associated with the new
Mathematics and Science Education Joint PhD Program between UCSD and San
Diego State University to train researchers in the cognitive study of
mathematics and science education. Level of appointment will be
determined by the qualifications of the applicant, but may be at any
level in the professorial series (assistant, associate, full).
University of California, San Diego
Math & Science Education Search Committee, EM
Department of Cognitive Science
9500 Gilman Drive
La Jolla, CA 92093-0515
ASSISTANT PROFESSOR COGNITIVE SCIENCE
The Department of Cognitive Science at the University of California, San
Diego seeks someone at the assistant professor level (tenure-track) who
can fulfill two important functions: (1) Represent a major area of
empirical research in human cognition; (2) Teach graduate and
undergraduate courses in experimental laboratory and field or
observational methods.
University of California, San Diego
Methods Search Committee, EM
Department of Cognitive Science
9500 Gilman Drive
La Jolla, CA 92093-0515
The following applies to both positions:
The Department takes a broadly based approach to topics in learning and
cognition, including experimental, observational, theoretical, and
computational investigations. Salary is commensurate with experience and
qualifications and based upon UC pay schedules. Women and minorities are
encourage to apply. The University of California, San Diego is an
affirmative action/equal opportunity employer.
First priority will be given to applications received by January 1, 1992.
Candidates should send a vita, reprints, a short letter describing their
background and interests, and names and addresses of at least three
references to the addresses listed above.
------------------------------
Subject: we want your messages!
From: issnnet@park.bu.edu
Date: Sun, 01 Sep 91 11:04:24 -0500
*** WE WANT YOUR POSTINGS !!!! ***
This is a reminder that the new USENET newsgroup comp.org.issnnet was
created earlier in August. The newsgroup is part of the International
Student Society for Neural Networks (ISSNNet) effort to establish a
centralized organization for students in Neural Networks.
Many of the messages posted to this group would be ideally suited for
posting to comp.org.issnnet. For example any requests for housing at a
conference site, or sharing rides, or inquiring about academic programs,
job offers, or even requests for references.
If you have no access to comp.org.issnnet you can still take advantage of
this newsgroup by e-mailing your message to "issnnet@park.bu.edu".
Please be sure to indicate you wish your message to be relayed to the
newsgroup, and include your name and e-mail address, and any other
information that may be needed for people to contact you.
Please take advantage of this new opportunity, and become part of this
useful new endeavour!
ISSNNet
------------------------------
Subject: NN s/w for 80x86 machines?
From: "David M. Miller" <miller@BTVLABVM.vnet.ibm.com>
Date: Thu, 26 Sep 91 14:55:30 -0500
[[ Editor's Note: At one point, I though of keeping a "canonical list"
but the field changes too rapidly for me. What is the current state of
the commercial art (ART3?) and/or public domain software for the PC? Can
anyone submit a recnt or up-to-date list. Thge last one I publishged was
February '91 or so. -PM ]]
Is there a comprehensive listing of neural networks that run on 80X86
platforms somewhere ?
Sincerely,
David M. Miller
IBM GTD
Dept. B07\ Bldg. 606-1
1000 River Rd.
Essex Junction, VT 05452-4299
(MILLER @ BTVLABVM)
(MILLER at BTVLABVM.VNET.IBM.COM)
(Phone 446-3414/2458)
------------------------------
Subject: New Journal and CFP
From: CVS15%CSPGCS11@CUNYVM.CUNY.EDU
Date: Fri, 27 Sep 91 06:29:18 -0800
[[ Editor's Note: This was sent in ALL CAPS. Any errors of mixed case are
therefore my own. I did not change grammar or spelling. -PM ]]
Information For All People Interested In NN
=--------------------------------------------------------------------------
Dear Coleagues: Using this opportunity we would like to iform you about
the new scientific journal devoted to neural networks, neurocomputing and
mass parallel systems called *Neural Network World* , which is since 1991
published by the idg co., Czecholovakia in Prague. This bimontly journal
intends to bring high level original contributions from the mentioned
area. Now three issues are already available. Between the members of the
international editorial board we have E.G.Prof.J.Taylor,
Prof.R.Hecht-Nielsen, Prof.H.Haken, Prof.V.Cimagalli, Prof.K.Hornik,
Prof.E .G.Kerckhoffs.
We would like to devote some of the coming issues in Spring 1992
especiall y to the problem how if at all! neural networks can contribute
to practical development of natural languages translation restricted to a
technical area etc! Therefore we will appreciate being submitted papers
concerning this field.
Of course, we are also interested in other contributions and in the
extension of the distribution of our journal.The IDG Co.,Czechoslovakia
is able to pay 15% profit share for each new subscriber.
Contact Address:
Dr.Mirko Novak, Institute Of Informatics And Computer
Science, Prague, Czechoslovakia FAX: (00422)8585789
------------------------------
Subject: Re: Welcome to Neuron-Digest
From: Kasra Hazeghi <kasra@bernina.ethz.ch>
Date: Wed, 02 Oct 91 15:33:54 +0700
[[ Editor's Note: Kasra is a new subscriber. Could some help with this
request, at least to get a person started. AT the same time, requests
like this really should be a bit more specific. What kind of images are
you looking at, for example? What papers have you seen which are
interesting? Remember, demonstrating that you have done a little
homework will result in much better responses! -PM ]]
I am scientific fellow at "communications technology Laboratory" of Swiss
Federal Institute of Technology and am interested in application of
neural networks to image coding and image compression. A colleage of mine
brought into my attention your Neuron Digest.
I would be gratefull for any information on recent papers or personal
experience on the application of neural networks for image data
compression.
Thank you for your assistance!
Kasra Hazeghi
------------------------------
Subject: Request for meeting in Hawaii
From: Humenberger Edmund <K11111I%EDVZ.UNI-Linz.AC.AT@CUNYVM.CUNY.EDU>
Date: Wed, 02 Oct 91 21:39:01 +0700
I got a information about a meeting in Hawaii in 92.
The theme is applikations of NN in Business.
Does anyone have information about it?
Thanks in advantage K11111i at alijku11.aearn
Humenberger Edmund
------------------------------
Subject: Request for info on neural nets and forecasting
From: Bob Snyder <snyderr@randb.abbott.com>
Date: Wed, 02 Oct 91 15:19:00 -0600
I am looking for some help from anyone who has successfully applied
neural networks to forecasting. I am looking at their use as an
alternative to the traditional methods of time series analysis (moving
average, exponential smoothing, adaptive filtering, etc.) and causal
techniques (correlation, regression, leading indicators, etc.). I would
like to look at business data (sales, market surveys) and use neural
networks as a strategic tool. Has someone used neural networks for
either time series or causal analysis? If so, could you provide some
references, neural network programs, datasets. Thanks for the help.
------------------------------
Subject: Re: Neuron Digest V8 #1
From: gcavelie <GCAVELIE%ANDESCOL@CUNYVM.CUNY.EDU>
Date: Wed, 02 Oct 91 17:33:51 -1100
[[ Editor's Note: I wasn't sure if this was private communication or
meant for publication. It had no other information or address. -PM ]]
To Oscar
About your subject on neural networks and caos, I am also interested,
perhaps we will be able to communicate and find bibliography in the
future.
GCAVELIE AT ANDESCOL.BITNET
------------------------------
Subject: IJCNN '91 Singapore - Request to share a room
From: Mark Evans <mre1@it-research-institute.brighton.ac.uk>
Date: Thu, 03 Oct 91 09:20:50 +0000
I will be attending IJCNN '91 in Singapore on the 18-21 November where I
will be presenting a paper. I would be interested in hearing from anyone
who would like to share a twin room for the duration of the conference.
(I am about to book myself a room or I could pay you if you have already
booked a room.) I am PhD student at Brighton Polytechnic, UK working in
the field of computer vision and neural networks.
Anyone interested ?
#################################################
# #
# M.R. Evans mre1@itri.bton.ac.uk #
# Research Assistant mre1@itri.uucp #
# #
# ITRI, #
# Brighton Polytechnic, #
# Lewes Road, #
# BRIGHTON, #
# E. Sussex, #
# BN2 4AT. #
# #
# Tel: +44 273 642915/642900 #
# Fax: +44 273 606653 #
# #
#################################################
------------------------------
Subject: Re: Review of Handwritten Character Recognition
From: mcsun!hp4nl!fwi.uva.nl!chris.fwi.uva.nl!smagt@uunet.uu.net (Patrick van der Smagt)
Organization: Faculteit Wiskunde & Informatica, Universiteit van Amsterdam
Date: 03 Oct 91 11:37:44 +0000
[[ Editor's Note: I'm not sure what this references, but it seems like a
good lead. -PM ]]
Try sending mail to ong@SPUNKY.CS.NYU.EDU. He wrote an
excellent report last year summarising many OCR techniques.
Patrick van der Smagt
------------------------------
Subject: Brain Design
From: "Raymond D. Scanlon" (CCB-R) <rscanlon@PICA.ARMY.MIL>
Date: Thu, 03 Oct 91 15:23:30 -0500
[[ Editor's Note: It seems Ray is looking mostly at single neuron
simulations, then combining them into higher order models. This is in
contrast to the more abstratc and mathematical models. Or have I misread
his message? In any case, I, for one, tend toward the biological. -PM ]]
I want to reach out to anyone designing or building a model of the
mammalian brain using neurons. I have a brain running in a Convex and am
putting one in a Connection Machine. The algorithms involved are those of
neural simulation.
Interests:
1. the scheme of connections between nuclei,
2. synaptic potentiation and depotentiation,
3. neural communication, including graded potential,
4. axonal growth,
5. hillock activity,
6. methods for showing the brain's actions.
Corollary interest:
1. the aware mind and the alert brain.
I hope there is someone out there with similar interests who would like
to communicate.
Ray Scanlon
------------------------------
Subject: walking robots
From: Thomas.Vogel@cl.cam.ac.uk
Date: Fri, 04 Oct 91 10:37:22 +0000
[[ Editor's Note: Again, a reminder that letting people know what you've
found already, before asking such a broad and general request, will help
others understand what you're looking for and also demonstrate that
you've done some initial "foot work." This is not to discourgae
"beginner's questions." If you are completely stumped as to how to begin
looking, just say so. If you need primers, explain what level you're
looking for. -PM ]]
I am currently working on a survey of recent activities in the field of
walking robots and especially robots that learn to walk. ANY pointer to
and comments about recent work in this field (using neural nets or other
algorithms) would be appreciated.
Thanks a lot
--thomas
==================================================================
Thomas U Vogel University of Cambridge
Computer Laboratory
Pembroke Street
email: tuv@cl.cam.ac.uk Cambridge CB2 3QG
tel : (+44) 223 334421 England
==================================================================
------------------------------
Subject: tree classification code available for comparative studies
From: Wray Buntine <wray@ptolemy.arc.nasa.gov>
Date: Fri, 04 Oct 91 16:11:01 -0800
I've made the following report available on the Neuroprose Archive
(cheops.cis.ohio-state.edu) as
buntine.treecode.ps.Z
not because I think connectionists are "deeply" interested in tree
learning research but because I think it would be a handy resource for
comparative studies:
1) systems such as CART/C4 are recognised programs
for benchmarking supervised learning systems against
2) home-grown reimplementations can be buggy and a timesink
3) if your problem has some inherent structure and a few
key indicator variables then trees may be a good thing to
try as well
4) trees typically don't work well with purely numeric
data or with problems with many variables all giving some
minor contribution to the prediction being made
The IND Tree Package we developed here incorporates some of early C4,
most of the classification trees component of CART (no regression) along
with some more recent Bayesian/MDL approaches that sometimes work better.
You can obtain LaTeX source for the following introductory report
if you email to:
ind@kronos.arc.nasa.gov
and ask for "About the IND Tree Package".
===============================================
About the IND Tree Package
Wray Buntine, RIACS
NASA Ames Research Center
Mail Stop 269-2
Moffet Field, CA 94035
September 29, 1991
This note introduces the IND Tree Package to prospective procurers and
those users/installers looking at IND for the first time.
IND does supervised learning using classification trees.
IND integrates features from Breiman {\it et al.}'s CART and Quinlan's C4
with newer Bayesian and minimum encoding methods for growing classification
trees, and provides an experimental control suite on top. The package comes
with a manual, ``man'' entries, and a guide to tree methods and research.
Information about obtaining IND, performance statistics,
documentation, authorship, copyright, installation, etc., are given.
IND is currently under development, although it has been used considerably
since late 1989. IND is implemented in C under UNIX.
Wray Buntine
RIACS (Research Inst. for Advanced Comp. Sc.)
NASA Ames Research Center phone: (415) 604 3389
Mail Stop 244-17 fax: (415) 604 6997
Moffett Field, CA, 94035 email: wray@ptolemy.arc.nasa.gov
------------------------------
Subject: Re: Neuron Digest V8 #1
From: "Hong Liang" <hliang@tera.cs.umn.edu>
Date: Sun, 06 Oct 91 19:56:54 -0600
> I am interested in having neural networks learn morphological operators
> like DILATION, EROSION, CLOSING, OPENING, SKELETONIZING.
This topic is often discussed in following Journals:
- IEEE Pattern Recognition
- IEEE Computer Vision
The major purpose is to find an optimal(quick/perfect/flexible) mothod
for complex graph processing.
Now I try other way.
Uncertain Information Representation Based On NN
- Input classification
- Datalog and problem domain generation
- Projection in Feature Map and Concept Graph
The set of methods can solve the problem of complex, fragment, and varied
graph identification.
Hong Liang
hliang@cs.umn.edu
------------------------------
End of Neuron Digest [Volume 8 Issue 2]
***************************************
