neuron-request@HPLABS.HP.COM (Neuron-Digest Moderator Peter Marvit) (09/27/88)
Neuron Digest Monday, 26 Sep 1988 Volume 4 : Issue 7 Today's Topics: Neural network simulator wanted Real Applications of Neural Networks ?? Colloquium on Neural Net Applications CONFERENCE NOTICE TSP in IEEE ICNN 88 in San Diego 1988 Tech Report Re: Music and Nets Send submissions, questions, mailing list maintenance and requests for back issues to "Neuron-request@hplabs.hp.com" ------------------------------------------------------------ Subject: Neural network simulator wanted From: djoslin@YEAGER.BBN.COM Date: Wed, 14 Sep 88 17:46:30 -0400 I've only become interested in neural networks recently, and attended the INNS conference in Boston last week. I was wondering if you know of a neural network simulator available for Unix. Preferably written in C and public domain. I grepped through the old comp.ai.neural-nets articles on our system and couldn't find any references. Thanks for any help. David [[ As previous issues of the Digest have mentioned, a number of companies are selling both software and hardware for doing ANN (Artifical Neural Net) simulations. One or small simulators made the rounds of ftp sites and comp.sources; I snarfed them but never tried them -- Hopfield nets, I think. I personally recommend going to your friendly bookstore and looking at PDP Vol 3 -- "Explorations" by Rumelhart and McClelland. There is both a PC version and a UNIX version! University of Rochester is also offering its simulator *for free* via ftp, or some nominal fee for tape and manual. It's getting close to the time to make another canonical list of simulators. If someone has such a beast, please let me know. Otherwise, just send me what you know and I'll collapse same into a future Digest. -PM]] ------------------------------ Subject: Real Applications of Neural Networks ?? From: Patrick Van Renterghem / Transputer Lab <PVR%BGERUG51.BITNET@CUNYVM.CUNY.EDU> Date: Thu, 15 Sep 88 09:48:00 +0100 Hello networkers, I am not a fanatic pro or con of connectionism and neural networks, but I am more interested in their applications than their basics. Your positive or negative results with these new techniques may swing my opinion. Replies will be used for a survey on the applicability of neural networks for real :-) applications. I have the following questions: * what kind of applications are neural networks used for ? I know pattern recognition is a favorite subject, and I would like to know more about specific realizations (and performance, compared to algorithic information processing), but there must be other application areas ??!!?? How about robotics, expert systems, image processing, ... * What coprocessor boards exist, what is the price, performance, their draw- backs, advantages, ... Addresses of manufacturers would be appreciated. Thanks in advance, Patrick Van Renterghem, State University of Ghent, Automatic Control Lab, The Transputer Lab div., Grotesteenweg Noord 2, B-9710 Ghent-Zwijnaarde, Belgium. P.S.: Companies listening and selling hard or software can send me information right away. [[ I will let the readership comment on the substance of Patrick's request. His return address works well. His is a common problem in this field; everyone has heard how great these Neural Nets are but a) what are they *good* for, b) what type of engineering can they be put to now, c) where can more information be found, d) what are the available hardware/software. These questions will resurface frequently in our nascent field. Many excellent resources exist to introduce people to the science of Neural Nets, but I'll admit ignorance for things to hand engineers who want to "use it now." I think the field is still too young and diffuse; we must learn from the cautionary tales from the early days of AI. Unfortunately, it's easy to be deluded by the relative ease by which NNs can produce results. At this point, my advise to people like Patrick is to be as *specific as possible *in your chosen application and search for related literature and other researchers or engineers who are working on similar applications. Secondly, try to attend one of the Neural Net conferences, especially those with vendor exhibits, and pester everyone for information. -PM]] ------------------------------ Subject: Colloquium on Neural Net Applications From: cook@Alliant.COM (Dale C. Cook) Date: Thu, 15 Sep 88 18:00:42 +0000 ANNOUNCING: A colloquium on Practical Neural Network Applications in Signal and Image Processing ORGANIZED BY: Ben Passarelli Alliant Computer Systems Company One Monarch Drive Littleton, MA 01460 WHERE: The Hyatt Regency, Burlingame, CA WHEN: Reception and meet the speakers: Sunday, October 30 Presentations: Monday and Tuesday, October 31 and Novemeber 1 8:30 am till 5:00 pm SPEAKERS AND TOPICS: Signal Processing and Neural Networks (KEYNOTE) Professor Bernard Widrow Department of Electrical Engineering Stanford University Neurons and Schemas: A Language for Visuomotor Coordination Professor Michael A. Arbib Director, Center for Neural Engineering University of Southern California Learning and Generalization in Connectionist Algorithms Professor David E. Rumelhart Department of Psychology Stanford University Cognitive Software for Neural Networks Professor James A. Anderson Department of Psychology Brown University Variation and Selection: An Evolutionary Learning Model for Signal Matching with Neural Networks Dr. Aviv Bergman SRI International Image Processing Using a Multi-Staged Neural Network Vision System Dr. Douglas Reilly Brown University and Nestor, Inc. Topic To Be Announced Professor Donald A. Glaser Nobel Laureate Department of Physics University of California Neural Network Dynamic Behavior Simulations for Vision Applications Dr. Ken Johnson Hughes Missile Systems Group Non-Linear Signal Processing with Neural Networks Dr. Alan Lapedes Theoretical Divison Los Alamos National Laboratories Physiology, Non-Linear Dynamics, Symmetries, and Self-Similar Structures in the Nervous System and Modeled Neural Networks Dr. Albert F. Lawrence Hughes Aircraft Company Images and Fuzzy Neural Networks Dr. Earl R. MacCormac Executive Director North Carolina Board of Science and Technology Image Processing Applications Dr. Robert Schudy The Analytical Sciences Corporation (TASC) ATR Neural Network Research on Speech Processing Dr. Yoh'ichi Tohkura Head of the Hearing and Speech Perception Department Advanced Telecommunications Research Institute International Panel Discussion on Commercializing Neural Network Applications Thomas J. Schwartz, Moderator Tom Schwartz Associates REGISTRATION: Before September 30: $445 After: $495 Registration fee includes all conferences, meals and breaks and the SF Bay cruise and reception, October 30. Registration deadline: October 15 CONTACT: Alliant Computer Systems Company One Monarch Drive Littleton, MA 01460 Ben Passarelli (408) 295-7222 or EMAIL: ben@alliant.alliant.com ....linus!alliant!ben Susan Ayers (508) 486-1454 or EMAIL: ayers@alliant.alliant.com ...linus!alliant!ayers - -- - Dale N1US VOICE: (617) 486-1343 ARPA: cook@alliant.alliant.com SMAIL: 1 Monarch Drive UUCP: ...linus!alliant!cook Littleton, MA 01460 ------------------------------ Subject: CONFERENCE NOTICE From: Julian_Dow%VME.GLASGOW.AC.UK@CUNYVM.CUNY.EDU Date: Thu, 15 Sep 88 18:10:43 -0000 ADVANCE NOTICE OF A TWO-DAY SESSION (APRIL 6th & 7th) ON BIOELECTRONICS & BIOSENSORS AS PART OF THE SPRING 1989 MEETING OF THE SOCIETY FOR EXPERIMENTAL BIOLOGY EDINBURGH, SCOTLAND : APRIL 3rd - 7th, 1989 ***Introduction The SEB provides a forum for British and European science across a broad spectrum of interest. The Spring 1989 meeting will contain a two-day session, on the 6th & 7th of April, on the impact of bioelectronics (defined as the area of overlap between cell biology, biochemistry, neurobiology and electronic engineering). The session is made possible by generous financial support from the SEB and the Scottish Development Agency (SDA). ***Topics covered Developmental biology: contact guidance, studies of cell behaviour on patterned surfaces of controlled topography. In vivo Neurobiology: measurement of electrical activity in intact nervous tissue with microengineered electrode arrays. In vitro Neurobiology: "real" neural networks of cultured neurons on arrays of electrodes. Relevance to computer design. Biomedical applications: use of electrode arrays as implantable sensors and in prosthetics. Biosensors: Fundamental problems. Application to biological problems in research. Technological problems: Choice of substrate, electrode materials. Long-term stability in aqueous environments. Multiplexing and signal treatment. Data reduction. Related sessions: Other sessions are planned on Computers in Neurobiology, and on Computers in Respiratory Physiology. ***Invited speakers Invited speakers include: Pickard (Cardiff), Turner (Cranfield), Gross (Texas), Edell (MIT), Pine (Caltech). ***Organisers Dr. J.A.T. Dow (Glasgow, Cell Biology) Prof. C.D.W. Wilkinson (Glasgow, Electronic Engineering) Dr. A.P.F. Turner (Cranfield, Bioelectronics) ***Format Plenary lectures, contributed oral papers and poster sessions. ***Venue The conference will be held on the campus of Edinburgh University, within easy reach of the historic centre of the city. Accommodation will be in Halls of Residence or in nearby hotels. Edinburgh is easily reached by local and international flights, by rail, or by road. It also provides an excellent starting point to visit other attractions in Scotland. ***Registration Pleasantly inexpensive, as the SEB is an educational charity. Registration will probably be around #50 for the conference. (This is not a registration form.) ***Interested? Please complete and return the attached form if you would like to be included on the mailing list, or write to Dr. Julian A.T. Dow, Department of Cell Biology, University of Glasgow, Glasgow G12 8QQ, Scotland. (Electronic mail address (JANET) : Julian_Dow@uk.ac.glasgow.vme) - ------------------------------------------------ SEB Conference: April 1989 To: Dr. Julian A.T. Dow, Department of Cell Biology, University of Glasgow, Glasgow G12 8QQ, Scotland. Telephone (041) 330 4616 Name: .............................................. Address: .............................................. .............................................. .............................................. .............................................. Please include me on your mailing list for further announcements on the Bioelectronics & Biosensors session of the 1989 SEB Spring Meeting in Edinburgh, on the 6th and 7th of April. I am interested in giving an oral presentation presenting a poster attending. My main area of interest is Cell biology Developmental Biology Neurobiology Biosensors Electronics Computing (Please circle) Please duplicate this form and pass on to any interested colleagues. ------------------------------ Subject: TSP in IEEE ICNN 88 in San Diego From: csrobe@work1.icase.edu (Charles S. [Chip] Roberson) Date: Fri, 16 Sep 88 10:59:34 -0400 During a poster session at INNS someone made a quick reference to a paper on Hopfield's implementation of the TSP that was presented at the IEEE International Conference on Neural Networks in San Diego. We have tried to get the proceedings but apparently they won't be available until later this year (though attendees already have their copies). If attendees do have copies already, would some kind sole please help us complete this reference so we can track down a copy of this paper. Also, if anybody has a on-line bibliography (e.g. BiBTeX) of any of the Neural Network conferences, I would really appreciate it if they would be willing to share. Thanks in advance for any replies to this ambiguous request (but that's what neural networks are for, right?) - -chip +-------------------------------------------------------------------------+ |Charles S. Roberson ARPANET: csrobe@icase.edu | |ICASE csrobe@[128.239.1.30] (cs.wm.edu) | |MS 132C BITNET: $csrobe@wmmvs.bitnet | |NASA Langley Rsch. Ctr. UUCP: ...!uunet!pyrdc!gmu90x!wmcs!csrobe| |Hampton, VA 23665-5225 Phone: (804) 865-4090 | +-------------------------------------------------------------------------+ ------------------------------ Subject: 1988 Tech Report From: jam@bu-cs.bu.edu (Jonathan Marshall) Date: Fri, 16 Sep 88 14:22:16 -0400 The following material is available as Boston University Computer Science Department Tech Report #88-010. It may be obtained from rmb@bu-cs.bu.edu or by writing to Regina Blaney, Computer Science Dept., Boston Univ., 111 Cummington St., Boston, MA 02215, U.S.A. I think the price is $7.00. - ----------------------------------------------------------------------- SELF-ORGANIZING NEURAL NETWORKS FOR PERCEPTION OF VISUAL MOTION Jonathan A. Marshall ABSTRACT The human visual system overcomes ambiguities, collectively known as the aperture problem, in its local measurements of the direction in which visual objects are moving, producing unambiguous percepts of motion. A new approach to the aperture problem is presented, using an adaptive neural network model. The neural network is exposed to moving images during a developmental period and develops its own structure by adapting to statistical characteristics of its visual input history. Competitive learning rules ensure that only connection ``chains'' between cells of similar direction and velocity sensitivity along successive spatial positions survive. The resultant self-organized configuration implements the type of disambiguation necessary for solving the aperture problem and operates in accord with direction judgments of human experimental subjects. The system not only accommodates its structure to long-term statistics of visual motion, but also simultaneously uses its acquired structure to assimilate, disambiguate, and represent visual motion events in real-time. - ------------------------------------------------------------------------ I am now at the Center for Research in Learning, Perception, and Cognition, 205 Elliott Hall, University of Minnesota, Minneapolis, MN 55414. I can still be reached via my account jam@bu-cs.bu.edu . --J.A.M. ------------------------------ Subject: Re: Music and Nets From: chrisley.pa@Xerox.COM Date: 18 Sep 88 22:50:00 -0700 Peter Marvit writes: [[ I heard that someone (Hinton?) demonstrated the production of Baroque-style fragments or phrases using a new architecture. I, too would be interested in this. -PM]] I believe you are referring to Prof. Kohonen's application of Dynamically Expanding Context, an algorithm for learning grammars, to the learning of Bach inventions. The description of the algorithm is found in the first paper of the second volume of the proceedings of the first IEEE ICNN conference. Since I just returned from a nine-month research stint with Prof. Kohonen, I could give (a few!) more details on the particular application and the algorithm if necessary. But be warned: those who are interested only in algorithms that can be explicitly linked to neural processes will not find this algorithm much "better" than learning paradigms in CS, AI, or Machine Learning (ala Valiant). Ron Chrisley Xerox Parc SSL New College 3333 Coyote Hill Rd Oxford OX1 3BN Palo Alto, CA 94304 England (415) 494-4740 0865-248451 [[ Thanks for the clarification; my recollection was from a conversation in passing. I appreciate the reference, though the application to music certainly sounded novel. -PM]] ------------------------------ End of Neurons Digest *********************