neuron-request@HPLABS.HP.COM ("Neuron-Digest Moderator Peter Marvit") (07/06/89)
Neuron Digest Wednesday, 5 Jul 1989 Volume 5 : Issue 28 Today's Topics: TR: CONNECTIONISM AND COMPOSITIONAL SEMANTICS EURASIP Workshop on Neural Networks TR available: Optimum Supervised Learning Connection Science Journal TRs on Reasoning and Somatagastic System Back Propagation vs Perceptrons (paper reference) Cognitive Science Society Meeting We're back on the air with lots of submissions in backlog! Send submissions, questions, address maintenance and requests for old issues to "neuron-request@hplabs.hp.com" or "{any backbone,uunet}!hplabs!neuron-request" ARPANET users can get old issues via ftp from hplpm.hpl.hp.com (15.255.16.205). ------------------------------------------------------------ Subject: TR: CONNECTIONISM AND COMPOSITIONAL SEMANTICS From: Dave.Touretzky@B.GP.CS.CMU.EDU Date: Wed, 31 May 89 21:53:16 -0400 CONNECTIONISM AND COMPOSITIONAL SEMANTICS David S. Touretzky School of Computer Science Carnegie Mellon University Pittsburgh, PA 15213-3890 Technical report CMU-CS-89-147 May 1989 Abstract: Quite a few interesting experiments have been done applying neural networks to natural language tasks. Without detracting from the value of these early investigations, this paper argues that current neural network architectures are too weak to solve anything but toy language problems. Their downfall is the need for ``dynamic inference,'' in which several pieces of information not previously seen together are dynamically combined to derive the meaning of a novel input. The first half of the paper defines a hierarchy of classes of connectionist models, from categorizers and associative memories to pattern transformers and dynamic inferencers. Some well-known connectionist models that deal with natural language are shown to be either categorizers or pattern transformers. The second half examines in detail a particular natural language problem: prepositional phrase attachment. Attaching a PP to an NP changes its meaning, thereby influencing other attachments. So PP attachment requires compositional semantics, and compositionality in non-toy domains requires dynamic inference. Mere pattern transformers cannot learn the PP attachment task without an exponential training set. Connectionist-style computation still has many valuable ideas to offer, so this is not an indictment of connectionism's potential. It is an argument for a more sophisticated and more symbolic connectionist approach to language. An earlier version of this paper appeared in the Proceedings of the 1988 Connectionist Models Summer School. ================ TO ORDER COPIES of this tech report: send electronic mail to copetas@cs.cmu.edu, or write the School of Computer Science at the address above. ------------------------------ Subject: EURASIP Workshop on Neural Networks From: Connectionists-Request@cs.cmu.edu Date: Thu, 01 Jun 89 10:13:45 -0400 CALL FOR PAPERS EURASIP WORKSHOP ON NEURAL NETWORKS Sesimbra, Portugal February 15-17, 1990 The workshop will be held at the Hotel do Mar in Sesimbra, Portugal. It will take place in 1990, from February 15 morning to 17 noon, and will be sponsored by EURASIP, the European Association for Signal Processing. It will be open to participants from all countries, both from inside and outside of Europe. Contributions from all fields related to the neural network area are welcome. A (non-exclusive) list of topics is given below. Care is being taken to ensure that the workshop will have a high level of quality. Proposed contributions will be evaluated by an international technical committee. A proceedings volume will be published, and will be handed to participants at the beginning of the workshop. The number of participants will be limited to 50. Full contributions will take the form of oral presentations, and will correspond to papers in the proceedings. Some short contributions will also be accepted, for presentation of ongoing work, projects (ESPRIT, BRAIN, DARPA,...), etc. They will be presented in poster format, and will not originate any written publication. A small number of non-contributing participants may also be accepted. The official language of the workshop will be English. TOPICS: - - signal processing (speech, image,...) - - pattern recognition - - algorithms (training procedures, new structures, speedups,...) - - generalization - - implementation - - specific applications where NN have been proved better than other approaches - - industrial projects and realizations SUBMISSION PROCEDURES: Submissions, both for long and for short contributions, will consist of (strictly) 2-page summaries. Three copies should be sent directly to the Technical Chairman, at the address given below. The calendar for contributions is as follows: Full contributions Short contributions Deadline for submission June 15, 1989 Oct 1, 1989 Notif. of acceptance Sept 1, 1989 Nov 15, 1989 Camera-ready paper Nov 1, 1989 ORGANIZING COMMITTEE General Chairman: Luis B. Almeida, INESC, Apartado 10105, P-1017 Lisboa, Codex, Portugal Phone: +351-1-544607; Fax: +351-1-525843; E-mail: {any backbone, uunet}!mcvax!inesc!lba Technical Chairman: Christian J. Wellekens, Philips Research Laboratory Brussels, Av. Van Becelaere 2, Box 8, B-1170 Brussels, Belgium Phone: +32-2-6742275; Fax: +32-2-6742299; E-mail: wlk@prlb2.uucp Technical committee: John Bridle (Royal Signal and Radar Establishment, Malvern, UK), Herve Bourlard (Intern. Computer Science Institute, Berkeley, USA), Frank Fallside (University of Cambridge, Cambridge, UK), Francoise Fogelman (Ecole de H. Etudes en Informatique, Paris, France), Jeanny Herault (Institut Nat. Polytech. de Grenoble, Grenoble, France), Larry Jackel (AT\&T Bell Labs, Holmdel, NJ, USA), Renato de Mori (McGill University, Montreal, Canada), H. Muehlenbein (GMD, Sankt Augustin, FRG). REGISTRATION, FINANCE, LOCAL ARRANGEMENTS: Joao Bilhim, INESC, Apartado 10105, P-1017 Lisboa, Codex, Portugal Phone: +351-1-545150; Fax: +351-1-525843. WORKSHOP SPONSOR EURASIP - European Association for Signal Processing CO-SPONSORS: INESC - Instituto de Engenharia de Sistemas e Computadores, Lisbon, Portugal IEEE, Portugal Section THE LOCATION: Sesimbra is a fishermens village, located in a nice region about 30 km south of Lisbon. Special transportation from/to Lisbon will be arranged. The workshop will end on a Saturday at lunch time; therefore, the participants will have the option of either flying back home in the afternoon, or staying for sightseeing for the remainder of the weekend in Sesimbra and/or Lisbon. An optional program for accompanying persons is being organized. ------------------------------ Subject: TR available: Optimum Supervised Learning From: Manoel Fernando Tenorio <tenorio@ee.ecn.purdue.edu> Date: Tue, 06 Jun 89 17:05:00 -0500 The Tech Report below will be available by June, 15. Please do not reply to this posting. Send all you requests to jld@ee.ecn.purdue.edu Self Organizing Neural Network for Optimum Supervised Learning Manoel Fernando Tenorio Wei-Tsih Lee School of Electrical Engineering School of Electrical Engineering Purdue University Purdue University W. Lafayette, IN. 47907 W. Lafayette, IN. 47907 tenorio@ee.ecn.purdue.edu lwt@ed.ecn.purdue.edu Summary Current neural network algorithms can be classified by the following characteristics: the architecture of the network, the error criteria used, the neuron transfer function, and the algorithm used during learning. For example: in the case of back propagation, one would classify the algorithm as a fixed architecture (feedforward in most cases), using a MSE criteria, and a sigmoid function on a weighted sum of the input, with the Generalized Delta Rule performing a gradient descent in the weight space. This characterization is important in order to assess the power of such algorithms from a modeling viewpoint. The expressive power of a network is intimately related with these four features. In this paper, we will discuss a neural network algorithm with noticeably different characteristics from current networks. The Self Organizing Neural Network (SONN) [TeLe88] is an algorithm that through a search process creates the network necessary and optimum in the sense of performance and complexity. SONN can be classified as follows. The network architecture is constructed through a search using Simulated Annealing (SA),and it is optimum in that sense. The error criteria used is a modification of the Minimum Description Length Criteria called the Structure Estimation Criteria (SEC); it takes into account both the performance of the algorithm and the complexity of the structure generated. The neuron transfer function is individually chosen from a pool of functions, and the weights are adjusted during the neuron creation. This function pool can be selected with a priori knowledge of the problem, or simply use a class of non-linearities shown to be general enough for a wide variety of problems. Although the algorithm is stochastic in nature (SA), we show that its performance is extremely high both in comparative and absolute terms. In [TeLe88], we have used SONN as an algorithm to identify and predict chaotic series, particularly the Mackey-Glass equation [LaFa87, Mood88] was used. For comparison, the experiments of using Back Propagation for this problem were replicated under the same computational environment. The results indicated that for about 10% of the computational effort, the SONN delivered a 2.11 times better model (normalized RMSE). Some inherited aspects of the algorithm are even more interesting: there were 3.75 times less weights, 15 times less connections, 6.51 times less epochs over the data set, and only 1/5 of the data was fed to the algorithm. Furthermore, the algorithm generates a symbolic representation of the network which can be used to substitute it, or be used for the analysis of the problem. ****************************************************************************** We have further developed the algorithm, and although not part of the report above, it will be part of a paper submitted to NIPS'89. There, some major improvements on the algorithm are reported. The same chaotic series problem can now run with 26.4 less epochs over the data set that BP, and have generated the same model in about 18.5 seconds of computer time. (This is down from 2 CPU hours in a Gould NP1 Powernode 9080). Performance on a Sun 3-60 was sightly over 1 minute. These performance figures include the use of an 8 times larger function pool; the final performance now independs of the size of the pool. Other aspects of the algorithm are also important considering. Because of its stochastic nature, no two runs of the algorithm should be the same. This can become a hindrance if a suboptimal solution is desired, since at every run the set of suboptimal models can be different. A report on modifications of the original SONN to run on an A* search are presented. Since the algorithm generates partial structures at each iteration, the learning process is only optimized for the structure presently generated. If such substructure is used as a part of a larger structure, then no provision is made to readjust its weights making the final model slightly stiff. A provision for melting the structure (parametric readjustment) is also discussed. Finally, the combination of symbolic processing with this numerical method can lead to construction of AI-NN based methods for supervised and unsupervised learning. The ability of SONN to take symbolic constraints and produce symbolic information can make such a system possible. Implications of this design are also explored. [LaFa87] - Alans Lapedes and Robert Farber, How Neural Networks Work, TR LA-UR-88-418, Los Alamos, 1987. [Mood88] - J. Moody, Fast Learning in Multi-Resolution Hierarchies, Advances in Neural Information Processing Systems, D. Touresky, Ed., Morgan Kaufmann, 1989 (NIPS88). [TeLe88] - M. F. Tenorio and W-T Lee, Self Organizing Neural Networks for the Identification Problem, Advances in Neural Information Processing Systems, D. Touresky, Ed., Morgan Kaufmann, 1989 (NIPS88). ------------------------------ Subject: Connection Science Journal From: Lyn Shackleton <lyn@CS.EXETER.AC.UK> Date: Wed, 07 Jun 89 13:36:25 -0000 ANNOUNCEMENT Issue 1. of the new journal CONNECTION SCIENCE has just gone to press and Issue 2. will follow shortly. The editors are very pleased with the response they have received and would welcome more high quality submissions or theoretical notes. VOLUME 1 NUMBER 1 CONTENTS Michael C Mozer & Paul Smolensky 'Using Relevance to Reduce Network Size Automatically' James Hendler 'The Design and Implementation of Symbolic Marker-Passing Systems' Eduardo R Caianello, Patrik E Eklund & Aldo G S Ventre 'Implementations of the C-Calculus' Charles P Dolan & Paul Smolensky 'Tensor Product Production System: A Modular Architecture and Representation' Christopher J Thornton 'Learning Mechanisms which Construct Neighbourhood Representations' Ronald J Williams & David Zipser 'Experimental Analysis of the Real-Time Recurrent Learning Algorithm' Editor: Dr NOEL E SHARKEY, Centre for Connection Science, Dept of Computer Science, University of Exeter, UK Associate Editors: Andy CLARK (University of Sussex, Brighton, UK) Gary COTTRELL (University of California, San Diego, USA) James A HENDLER (University of Maryland, USA) Ronan REILLY (St Patrick's College, Dublin, Ireland) Richard SUTTON (GTE Laboratories, Waltham, MA, USA) FORTHCOMING IN VOLUMES 1 & 2 Special Issue on Natural Language, edited by Ronan Reilly & Noel Sharkey Special Issue on Hybrid Symbolic/Connectionist Systems, edited by James Hendler For further details please contact. lyn shackleton (assistant editor) Centre for Connection Science JANET: lyn@uk.ac.exeter.cs Dept. Computer Science University of Exeter UUCP: !ukc!expya!lyn Exeter EX4 4PT Devon BITNET: lyn@cs.exeter.ac.uk.UKACRL U.K. ------------------------------ Subject: TRs on Reasoning and Somatagastic System From: Ron Sun <rsun@cs.brandeis.edu> Date: Mon, 12 Jun 89 09:07:57 -0400 [[ Editor's Note: This is an interesting juxtaposition of topics! Wet-ware enthusiasts, I'm sure, would wish to remind the non-biologists in the field that the spiny lobster stomatogastric system is one of the best understood complete "neural nets" we currently have (another being, the sluggish aplysia). Creating computational models which provide predictions, not merely mimicry, would be a great step forward in the wedding of artifical and natural neural network models. -PM ]] The following two tech reports are available from rsun%cs.brandeis.edu@relay.cs.net or R. Sun Brandeis U. CS Waltham, MA 02254 ############################################################# A Discrete Neural Network Model for Conceptual Representation and Reasoning Ron Sun Computer Science Dept. Brandeis University Waltham, MA 02254 Current connectionist models are oversimplified in terms of the internal mechanisms of individual neurons and the communication between them. Although connectionist models offer significant advantages in certain aspects, this oversimplification leads to the inefficiency of these models in addressing issues in explicit symbolic processing, which is proven to be essential to human intelligence. What we are aiming at is a connectionist architecture which is capa- ble of simple, flexible representations of high level knowledge structures and efficient performance of reasoning based on the data. We first propose a discrete neural net- work model which contains state variables for each neuron in which a set of discrete states is explicitly specified instead of a continuous activation function. A technique is developed for representing concepts in this network, which utilizes the connections to define the concepts and represents the concepts in both verbal and compiled forms. The main advantage is that this scheme can handle variable bindings efficiently. A reasoning scheme is developed in the discrete neural network model, which utilizes the inherent parallelism in a neural network model, performing all possible inference steps in parallel, implementable in a fine-grained massively parallel computer. (to appear in Proc. CogSci Conf. 1989) ############################################################### Model local neural networks in the lobster stomatogastric ganglion Ron Sun Eve Marder David Waltz Brandeis University Waltham, MA 02254 ABSTRACT We describe a simulation study of the pyloric network of the lobster stomatogastric ganglion. We demonstrate that a few simple activation functions are sufficient to describe the oscillatory behavior of the network. Our aim is to determine the essential mechanisms necessary to specify the operation of biological neural networks so that we can incorporate them into connectionist models. Our model includes rhythmically active bursting neurons and long time-constant synaptic relations. In the process of doing this work, various models and algorithms were compared. We have derived some connectionist learning algorithms. They have proved useful in terms of ease and accuracy in model generation. (to appear in IJCNN-89) ************************************************************** ------------------------------ Subject: Back Propagation vs Perceptrons (paper reference) From: worden@emx.utexas.edu (Sue J. Worden) Date: Tue, 13 Jun 89 06:09:50 -0500 Persons working with "structured" type neural networks may be interested in the following paper: Brady, Martin L.; Raghu Raghavan; Joseph Slawny. "Back Propagation Fails to Separate Where Perceptrons Succeed". IEEE Trans. Circuits & Systems, Vol.36, No.5, May 1989, pp.665-674. Abstract: "It is widely believed that the back propagation algorithm in neural networks, for tasks such as pattern classification, overcomes the limitations of the perceptron. We construct several counterexamples to this belief. We also construct linearly separable examples which have a unique minimum which fails to separate two families of vectors, and a simple example with four two-dimensional vectors in a single layer network showing local minima with a large basin of attraction. Thus back propagation is guaranteed to fail in the first, and likely to in the second, example. We show that even multilayered (hidden layer) networks can also fail in this way to classify linearly separable problems. Since our examples are all linearly separable, the perceptron would correctly classify them. Our results disprove the presumption, made in recent years, that, barring local minima, back propagation will find the best set of weights for a given problem." ------------------------------ Subject: Cognitive Science Society Meeting From: Sue Schuon <sue@csmil.umich.edu> Date: Fri, 09 Jun 89 15:24:18 -0400 ********************************* * Program Announcement * *The 11th Annual Conference of * *The Cognitive Science Society * * August 16-19, 1989 * * University of Michigan * * Ann Arbor, Michigan * ********************************* TECHNICAL PROGRAM INVITED SPEAKERS Roger Shepard, Stanford University Ulric Neisser, Emory University David Waltz, Thinking Machines SYMPOSIA Cognitive Science Ten Years After the Society's Initial Meeting Distributed Representations in Adaptive Systems Connectionist Models of Language Conceptual Change in Scientists and Children The Role of Attention in High Level Vision: Cognitive Neuroscience Perspectives Rules & Inductive Reasoning Distributed Cognition On-Line Analyses of Sentence Processing: The Role of Structure and Inference Applications of Cognitive Science TECHNICAL PAPERS POSTER SESSIONS PRE-CONFERENCE TUTORIALS Parallel Distributed Processing, Jay McClelland & David Rumelhart (Full Day) Soar, John Laird, Alan Newell, & Paul Rosenbloom (Half Day) GENERAL INFORMATION SCHEDULE Wednesday 9:00 am - 5:00 pm -- PDP Tutorial 1:00 pm - 5:00 pm -- Soar Tutorial 7:00 pm -10:30 pm -- Opening Reception & Poster Session Thursday 8:30 am - 6:00 pm -- Technical Program 7:00 pm -10:00 pm -- Reception & Poster Session Friday 8:30 am - 6:00 pm -- Technical Program 6:30 pm - 9:00 pm -- Banquet Saturday 8:00 am - 5:00 pm -- Technical Program LOCATION The 11th Annual Conference of the Cognitive Science Society will be held in Ann Arbor, Michigan, on the Central Campus of The University of Michigan. Conference sessions will be held in the Michigan League, 911 N. University. The meeting is located within walking distance of restaurants, shopping areas, museums and galleries. Even during the summer the University offers a wide range of cultural activities including plays, concerts, libraries, and movies. Ann Arbor also provides a rich mixture of recreational and cultural events. In some of the excellent - parks, for example, you may rent equipment for canoeing and paddle boating or go swimming or take a botanical garden walk. ADVANCED REGISTRATION To assure your reservation for the conference, please complete the accompanying registration form and return it to the address below: University of Michigan Extension Service Department of Conferences & Institutes 200 Hill Street Ann Arbor, MI 48104-3297 The conference registration desk will open at 7:30 a.m. on Wednesday, August 16, on the second floor Concourse of the Michigan League, 911 N. University Avenue. It will remain open during the tutorials and reopen at 7:30 a.m. on Thursday, August 17, for the conference sessions. Staff will be available throughout the conference at the registration desk. FEES Conference Registration (August 17 - 19) Advanced (by Aug. 1) On-site (after Aug. 1) Member: $120.00 $160.00 Nonmember: $160.00 $200.00 Student: $80.00 $100.00 Banquet: $30.00 Tutorial Registration (August 16) PDP Soar University/Government: $120.00 $ 60.00 Corporate: $250.00 $125.00 Student: $ 80.00 $ 40.00 Fees for advanced registration by check or money order payable to University of Michigan Extension Service. Remittance should be made in U.S. funds drawn on a U.S. bank. Advance registrants' conference fees may be charged on VISA and Mastercard as indicated on the registration form. Credit cards will not be accepted for on- site registration. The banquet has a separate fee and should be ordered in advance. The Conference Banquet will be at the Michigan League on Friday evening, August 18, and the fee is $30 per person. Please send payment for banquet together with the registration fee. CANCELLATIONS Cancellations received by telephone or written application prior to August 9 will be entitled to a full refund minus a $10 handling charge. Refunds will take approximately six to eight weeks. Those who do not appear at the conference, thereby cancelling by default, will not be eligible to receive any refund. CONFIRMATION OF REGISTRATION Acknowledgements confirming registration in the conference sessions, guests' programs, etc, will be sent to the participants after the remittance has been received if time permits. Off-campus lodging confirmations will be sent directly from the hotel chosen by the participant. ACCOMMODATIONS Accommodations for participants in the conference and their spouses and guests are available at the facilities listed below. To arrange for hotel rooms, please contact the hotel directly. Reservations should be placed not later than July 25, 1989, to assure accommodations. These facilities will honor reservations until 6:00 p.m. on the date you indicate you will arrive. Reservations will be held after that time only if prior arrangements are made to guarantee late arrival. All rates are subject to change and a 6% tax is levied. Mention the Cognitive Science Society when making your reservation. Ann Arbor Inn: Huron at Fourth Avenue, Ann Arbor, MI 48104, (313) 769-9500, about four blocks from the conference. Rates: $62, single occupancy, and $72, double occupancy. Bell Tower Hotel: 300 S. Thayer Street, Ann Arbor, MI 48104, (313)769-3010, less than one block from the conference. Rates: $70, single occupancy, and $80, double occupancy. Campus Inn: East Huron at State Street, Ann Arbor, MI 48104, (313)769-2200, about two blocks from the conference. Rates: $60, single occupancy, and $70, double occupancy. University Residence Hall Lodging: Mosher-Jordan Residence Hall, three blocks northeast of the meeting on U-M Central Campus, and Cambridge House, three blocks southwest, will offer lodging the nights of August 15-18. All rooms have single beds or bunk beds; there are no double beds. In Mosher-Jordan, participants will share several large communal bathrooms on each floor. Cambridge House features private bath facilities. Both Residence Halls are air-conditioned. Residence Hall front desk hours are 7:30 a.m. to 11:00 p.m. Only Residence Hall lodgers will have keys and access to the buildings 24 hours a day. Cost for rooms in both residence halls is $108.16, single, and a double is $74.88 per person for the four-night package. Participants with accompanying children under 12 are not encouraged to stay in the dormitories, due to the lack of space and lounge areas. Children 12 and over may stay in the dorms at the adult rate. Participants desiring lodging in the residence halls should make reservations on the Conference Registration form. Payment for residence hall lodging should be made in advance with the registration fees. No residence hall reservations will be accepted after July 25, 1989. If residence hall reservations are cancelled after August 1, cost of a night's lodging will be deducted from the refund. The residence hall package provides lodging for the nights of August 15-18 (check-out by noon on August 19). If you need lodging prior to the 15th or after the l9th, you must make arrangements with a local hotel or there may be space available in the Mary Markley Residence Hall, 1503 Washington Hts. (across the street from Mosher- Jordan/Stockwell). To stay at Mary Markley (313)764-5297, you will need to make your own reservation and make payment directly to The University of Michigan Housing Office. CHILD CARE Due to insurance constraints, The University cannot provide child care. The hotels have some child care arrangements, but must be contacted in advance by the participant to discuss details. MEALS A banquet open to all registrants and their spouses and guests will be held at the Michigan League on Friday, August 18, 1989. The cost of the banquet is $30 per person. Vegetarians and persons with other special dietary/medical requirements should indicate that information on the conference registration form. All other meals may be obtained on your own at restaurants and hotels in the Ann Arbor area. There are numerous restaurants within five blocks of the conference. Ethnic specialties, fast food, cafeteria fare, and excellent American dishes can be chosen. Costs range from economical to expensive; ambiance, from spartan to luxurious. Listings will be provided among the conference materials distributed at the meeting. TRANSPORTAION Ann Arbor may be reached by car via US-23 from the north and south and I-94 and M-14 from the east and west. The city is served by the Detroit Metropolitan Airport (25 miles east of Ann Arbor) with regular daily flights via major domestic and international airlines. There is regularly scheduled limousine service between the airport and Ann Arbor. Tickets may be purchased at the Ground Transportation Desk at the airport for $13 one-way and $24 round trip. There is a "no refund" policy. Stops will be made at the conference hotels and University residence halls. Return reservations are required; call 1-800- 351-5466 from Ann Arbor. Taxi service from the airport to Ann Arbor hotels is approximately $30.00 one way. Major car rental companies are located at the Detroit Metropolitan Airport. Ann Arbor is also served by Greyhound Bus and Amtrak Railroad Service. Participants are urged to contact their local travel offices for current air, bus, and rail service schedules and rates. Most conference activities are within comfortable walking distance of one another. There is no need to bring a car to this event unless you have additional plans which would require independent transportation. Ann Arbor has an excellent bus system which provides access to the city and surrounding communities. Each ride is 60 cents. University buses are free and serve all campus areas. SOCIAL EVENTS A welcoming reception and poster session will be held in the Michigan League Ballroom on Wednesday, August 16, from 5-7 p.m. A cash bar and light hors d'oeuvres will be provided. On Thursday, August 17, there will be a reception in the Ballroom in conjunction with the poster display. These events will provide opportunities to greet old friends and welcome new Society members. On Friday, August 18, the Society will hold its banquet from 6:30-9 p.m. at the Michigan League. The cost of the event is $30 per person. Tickets may be ordered on the registration form. Participants would be well advised to order banquet ticket(s) in advance. RECREATION Conference participants and their registered guests may purchase passes to use the Central Campus Recreation Building (CCRB), located near the Residence Halls, two blocks from the conference sites. A weekly pass costs $12 and a daily pass is $4 per person. Passes can be purchased at the conference registration desk or at the Residence Hall registration desk upon arrival. CCRB contains courts for racquetball, squash, volleyball and basketball, and a swimming pool, indoor jogging track, weight training rooms, saunas, and other features. Nearby outdoor facilities include a jogging track, softball fields and tennis courts. SMOKING POLICY No smoking is allowed in University buildings except in designated areas. The designated areas are identified in public buildings and you may smoke in your own room in the Residence Hall. This will mean that there is no smoking allowed in the meeting rooms where the conference will be held. PARKING Limited parking is available in the University's structures on Fletcher Street, one block east of the Rackham Building; on Thayer Street, one block west of the Rackham Building; and on Thompson Street, one block west of the Michigan Union, at $4.00 per day, payable at the structures. Participants staying in Mosher-Jordan and Cambridge House Residence Halls may buy parking permits for a University parking facility at $4.00 per day. These permits will be available at the main desk in the dormitory. Such parking is usually five-six blocks from the Residence Halls. Conference hotels provide different parking arrangements for registered guests. Please check with the hotel when making your original reservations to determine the policy. WEATHER Ann Arbor is a northern city, some 55 miles from Lake Erie. Average August temperatures range from nighttime lows of 57 F degrees to daytime highs of 79 F degrees. Rain, although infrequent, is possible. Further Information Additional information about registration and other administrative questions should be addressed to: The University of Michigan Extension Service Department of Conferences & Institutes 200 Hill Street Ann Arbor, MI 48104-3297 phone: (313) 764-5304 Requests or questions concerning the conference technical program should be addressed to Gary M. Olson, University of Michigan, Cognitive Science and Machine Intelligence Laboratory, 701 Tappan Street, Ann Arbor, MI 48109-1234 or gmo@csmil.umich.edu ************************************************** ************************************************** ** 11th Annual Conference of the ** ** Cognitive Science Society ** ** The University of Michigan ** ** Ann Arbor, Michigan ** **August 16-19, 1989 Registration Form 401684** ************************************************** ************************************************** ********************************************************* *(for office use only) ck/m.o.# ______$ ______________ * *Personal check ____ Other check ____ * *Issued by ________________________________ * ********************************************************* Please complete this form and return before August 1, 1989, to assure acknowledgement. Accompanying visitors should be included on participant's form. Please print or type. Name _______________________ Title ___________________ Organization _________________________________________ Address _____________________________________________ City ____________ State _______ Zip ____________ Daytime Phone ( )___________ E-mail ____________ Accompanied by (Names, include ages of children) ______________________________________________________ ______________________________________________________ **ADVANCED REGISTRATION FEES** _____ Member @ $120.00 $_________________ _____ Nonmember @ $160.00 $_________________ _____ Student @ $ 80.00 $_________________ _____ Banquet @ $ 30.00 $_________________ Special Dietary Requirement _____________________ **TUTORIAL FEES** Parallel Distributed Processing _____ University/Government @ $120.00 $___________ _____ Corporate @ $250.00 $___________ _____ Student @ $80.00 $___________ Soar _____ University/Government @ $60.00 $___________ _____ Corporate @ $125.00 $___________ _____ Student @ $40.00 $___________ *RESIDENCE HALL HOUSING* (fees are per person and include 4 nights) _____ Single @ $108.16 $___________________ _____ Double @74.88 $___________________ Circle one: Male Female Dorm-arrange roommate? Yes _____ No _____ Name of pre-arranged roommate: ____________________ Total Fees $_____________________ *METHOD OF PAYMENT* _____ Enclosed is check/money order for the total amount in US funds payable to University of Michigan Extension Service. _____ Please charge my credit card#__________________ _____ VISA _____ Mastercard Expiration date: ________ Signature: _______________________________ MAIL TO: U-M Extension Service, Conferences & Institutes, 200 Hill Street, Ann Arbor, MI 48104-3297. Telephone: (313) 764-5304 ------------------------------ End of Neurons Digest *********************