ADVANCES IN 
NEURAL 
INFORMATION 
PROCESSING 
SYSTEMS 3 
Other Titles of Interest from 
Morgan Kaufmann Publishers 
NIPS 
-2- 
Advances in Neural Information Processing Systems 
Proceedings of the 1989 Conference 
Edited by David S. Touretzky 
NIPS -1- 
Advances in Neural Information Processing Systems 
Proceedings of the 1988 Conference 
Edited by David S. Touretzky 
Computer Systems That Learn: Classification and Prediction Methods from 
Statistics, Neural Nets, Machine Learning, and Expert Systems 
By Sholom M. Weiss and Casimir A. Kulikowski 
Connectionist Models Summer School Proceedings 
1990 Edited by David S. Touretzky, Jeffrey L. Elman, Terrence J. 
Sejnowski, and Geoffrey E. Hinton 
1988 Edited by David S. Touretzky, Geoffrey E. Hinton, and 
Terrence J. Sejnowski 
Learning Machines 
By Nils J. Nilsson, with an Introduction by Terrence Sejnowski and 
Halbert White 
Readings in Speech Recognition 
Edited by Alex Waibel and Kai-Fu Lee 
COLT Proceedings of the Annual Workshops on Computational Learning 
Theory: 
1990 Edited by Mark Fulk and John Case 
1989 Edited by Ron Rivest, Manfred Warmuth, and David Haussler 
1988 Edited by David Haussler and Leonard Pitt 
Genetic Algorithms: Proceedings of the Third International Conference 
Edited by David Schaffer 
Readings in Cognitive Science: A Perspective from Psychology 
Intelligence 
Edited by Allan Collins and Edward Smith 
and Artificial 
ADVANCES IN 
NEURAL 
INFORMATION 
PROCESSING 
SYSTEMS 3 
EDITED BY 
RICHARD P. LIPPMANN 
MIT LINCOLN LABORATORY 
JOHN E. MOODY 
YALE UNIVERSITY 
DAVID S. TOURETZKY 
CARNEGIE MELLON UNIVERSITY 
MORGAN KAUFMANN PUBLISHERS 
2929 C,vtPUS DRIVE 
SUITE 260 
SAN MATEO, CALIFORNIA 94403 
Editor Bruce M. Spatz 
Production Editor Sharon Montooth 
Cover Design Jo Jackson 
Compositor Technically Speaking Publications 
CIP data is available. 
ISSN 1049-5258 
ISBN 1-55860-184-8 
MORGAN KAUFMANN PUBLISHERS, INC. 
2929 Campus Drive 
San Mateo, CA 94403 
(415) 578-9911 
 1991 by Morgan Kaufmann Publishers, Inc. 
All rights reserved. 
Printed in the United States of America 
No part of this publication may be reproduced, stored in a retrieval sys- 
tem, or transmitted in any form or by any means electronic, mechani- 
cal, photocopying, recording, or otherwisewithout the prior permis- 
sion of the publisher. 
95 94 93 92 91 5 4 3 2 1 
Contents 
Part I Neurobiology 
Studies of a Model for the Development and Regeneration of Eye-Brain Maps . . 
J.D. Cowan and A.E. Friedman 
Development and Spatial Structure of Cortical Feature Maps: A Model Study . . 
K. Obermayer, H. Ritter, and K. Schulten 
Interaction Among Ocularity, Retinotopy and On-center/Off-center Pathways . . 
Shigeru Tanaka 
Simple Spin Models for the Development of Ocular Dominance Columns and Iso- 
Orientation Patches ........................... 
J.D. Cowan and A.E. Friedman 
A Recurrent Neural Network Model of Velocity Storage in the Vestibulo-Ocular 
Reflex ................................. 
Thomas J. Anastasio 
Self-organization of Hebbian Synapses in Hippocampal Neurons ........ 
Thomas H. Brown, Zachary F. Mainen, Anthony M. Zador, and Brenda J. Claiborne 
Cholinergic Modulation May Enhance Cortical Associative Memory Function . . 
Michael E. Hasselmo, Brooke P. Anderson, and James M. Bower 
Part II Neuro-Dynamics 
Order Reduction for Dynamical Systems Describing the Behavior of Complex 
Neurons ................................ 
Thomas B. Kepler, L.F. Abbott, and Eve Marder 
Stochastic Neurodynamics ........................ 
J.D. Cowan 
Dynamics of Learning in Recurrent Feature-Discovery Networks ........ 
Todd K. Leen 
A Lagrangian Approach to Fixed Points ................... 
Eric Mjolsness and Willard L. Miranker 
Associative Memory in a Network of 'Biological' Neurons .......... 
Wulfram Gerstner 
CAM Storage of Analog Patterns and Continuous Sequences with 3N 2 Weights . 
Bill Baird and Frank Eeckman 
3 
11 
18 
26 
32 
39 
46 
55 
62 
70 
77 
84 
91 
v 
Connection Topology and Dynamics in Lateral Inhibition Networks 
CdVI. Marcus, F.R. Waugh, and R.M. Westervelt 
Shaping the State Space Landscape in Recurrent Networks ....... 
Patrice Y. Simard, Jean Pierre Raysz, and Bernard Victorri 
Adjoint-Functions and Temporal Learning Algorithms in Neural Networks 
N. Toomarian and ]. Barhen 
Part III Oscillations 
Phase-coupling in Two-Dimensional Networks of Interacting Oscillators . . . 
Ernst Niebur, Daniel M. Kammen, Christof Koch, Daniel Ruderman, and Heinz G. 
Schuster 
Oscillation Onset in Neural Delayed Feedback . 
Andr Longtin 
Analog Computation at a Critical Point .... 
Leonid Kruglyak and William Bialek 
Part IV Temporal Reasoning 
Modeling Time Varying Systems Using Hidden Control Neural Architecture 
Esther Levin 
The Tempo 2 Algorithm: Adjusting Time-Delays By Supervised Learning .... 
Ulrich Bodenhausen and Alex Waibel 
A Theory for Neural Networks with Time Delays 
Bert de Vries and Jose C. Principe 
ART2/BP Architecture for Adaptive Estimation of Dynamic Processes 
Einar Sorheim 
Statistical Mechanics of Temporal Association in Neural Networks 
Andreas V.M. Herz, Zhaoping Li, and ]. Leo van Hemmen 
Learning Time-varying Concepts .............. 
Anthony Kuh, Thomas Petsche, and Ronald L. Rivest 
The Recurrent Cascade-Correlation Architecture ........ 
Scott E. Fahlman 
98 
105 
113 
123 
130 
137 
147 
155 
162 
169 
176 
183 
190 
Part V Speech 
Continuous Speech Recognition by Linked Predictive Neural Networks .... 
Joe Tebelskis, Alex Waibel, Bojan Petek, and Otto Schmidbauer 
A Recurrent Neural Network for Word Identification from Continuous Phoneme 
Strings ................................ 
Robert B. Allen and Candace A. Kamm 
 199 
vi 
Connectionist Approaches to the Use of Markov Models for Speech Recognition . 
Herv Boutlard, Nelson Morgan, and Chuck Woofers 
Spoken Letter Recognition ........................ 
Mark Fant and Ronald Cole 
Speech Recognition Using Demi-Syllable Neural Prediction Model ....... 
Ken-ichi Iso and Takao Watanabe 
RecNorm: Simultaneous Normalisation and Classification Applied to Speech 
Recognition ............................... 
John S. Bridle and Stephen J. Cox 
Exploratory Feature Extraction in Speech Signals ............... 
Nathan Intrator 
Phonetic Classification and Recognition Using the Multi-Layer Perceptton .... 
Hong C. Leung, James R. Glass, Michael S. Phillips, and Victor W. Zue 
From Speech Recognition to Spoken Language Understanding ......... 
Victor Zue, James Glass, David Goodine, Lynette Hirschman, Hong Leung, Michael 
Phillips, Joseph Polifroni, and Stephanie Seneft 
Speech Recognition using Connectionist Approaches ............. 
Khalid Choukri 
Part VI Signal Processing 
Natural Dolphin Echo Recognition Using an Integrator Gateway Network... 
Herbert L. Roitblat, Patrick W.B. Moore, Paul E. Nachtigall, and Ralph H. Penner 
Signal Processing by Multiplexing and Demultiplexing in Neurons ...... 
David C. Tam 
Applications of Neural Networks in Video Signal Processing 
John C. Pearson, Clay D. Spence, and Ronald Sverdlove 
Part VII Visual Processing 
Discovering Viewpoint-Invariant Relationships That Characterize Objects 
Richard S. Zemel and Geoffrey E. Hinton 
A Neural Network Approach for Three-Dimensional Object Recognition 
Volker Tresp 
A Second-Order Translation, Rotation and Scale Invariant Neural Network 
Shelly D.D. Goggin, Kristina M. Johnson, and Karl E. Gustafson 
Learning to See Rotation and Dilation with a Hebb Rule ........ 
Martin I. Sereno and Margaret E. Sereno 
Stereopsis by a Neural Network Which Learns the Constraints ..... 
Alireza Khotanzad and Ying-Wung Lee 
213 
220 
227 
234 
241 
248 
255 
262 
273 
282 
289 
299 
306 
313 
320 
327 
vii 
Grouping Contours by Iterated Pairing Network ............... 
Amnon Shashua and Shimon Ullman 
Neural Dynamics of Motion Segmentation and Grouping ........... 
Ennio Mingolla 
A Multiscale Adaptive Network Model of Motion Computation in Primates . . . 
H. Taichi Wang, Bimal Matbur, and Christof Koch 
Qualitative Structure From Motion ..................... 
Daphna Weinshall 
Optimal Sampling of Natural Images .................... 
William Bialek, Daniel L. Ruderman, and A. Zee 
A VLSI Neural Network for Color Constancy ................ 
Andrew Moore, John Allman, Geoffrey Fox, and Rodney Goodman 
Optimal Filtering in the Salamander Retina ................. 
Fred Rieke, W. Geoffrey Owen, and William Bialek 
A Four Neuron Circuit Accounts for Change Sensitive Inhibition in Salamander 
Retina ................................. 
Jeffrey L. Teeters, Frank H. Eeckman, and Frank S. Werblin 
Feedback Synapse to Cone and Light Adaptation ............... 
Josef Skrzypek 
An Analog VLSI Chip for Finding Edges from Zero-crossings ......... 
Wyeth Bair and Christof Koch 
A Delay-Line Based Motion Detection Chip ................. 
Tim Horiuchi, John Lazzaro, Andrew Moore, and Christof Koch 
Part VIII Control and Navigation 
Neural Networks Structured for Control Application to Aircraft Landing .... 
Charles Schley, Yves Chauvin, Van Henkle, and Richard Golden 
Real-time Autonomous Robot Navigation Using VLSI Neural Networks .... 
Lionel Tarassenko, Michael Brownlow, Gillian Marshall, Jon Tombs, and Alan 
Murray 
Rapidly Adapting Artificial Neural Networks for Autonomous Navigation . . . 
Dean A. Pomerleau 
Learning Trajectory and Force Control of an Artifidal Muscle Arm ....... 
Masazumi Katayama and Mitsuo Kawato 
Proximity Effect Corrections in Electron Beam Lithography .......... 
Robert C. Frye, Kevin D. Cummings, and Edward A. Reitman 
335 
342 
349 
356 
363 
370 
377 
384 
391 
399 
406 
415 
422 
429 
436 
443 
VIII 
Planning with an Adaptive World Model .................. 
Sebastian B. Thrun, Knut Mi)'ller, and Alexander Linden 
A Connectionist Learning Control Architecture for Navigation ......... 
Jonathan R. Bachrach 
Navigating Through Temporal Difference .................. 
Peter Dayan 
Integrated Modeling and Control Based on Reinforcement Learning ...... 
Richard S. Sutton 
A Reinforcement Learning Variant for Control Scheduling ........... 
Aloke Guha 
Adaptive Range Coding ......................... 
Bruce E. Rosen, James M. Goodwin, and Jacques J. Vidal 
Neural Network Implementation of Admission Control ............ 
Rodolfo A. Milito, Isabelle Guyon, and Sara A. Solla 
Reinforcement Learning in Markovian and Non-Markovian Environments . . . 
fiirgen Schmidhuber 
A Model of Distributed Sensorimotor Control in The Cockroach Escape Turn . . 
R.D. Beer, G.J. Kacmarcik, R.E. Ritzmann, and H.J. Chiel 
Flight Control in the Dragonfly: A Neurobiological Simulation ......... 
William E. Faller and Marvin W. Luttges 
Part IX Applications 
A Novel Approach to Prediction of the 3-Dimensional Structures ........ 
Henrik Fredholm, Henrik Bohr, Jakob Bohr, Soren Brunak, Rodney M.J. Cotterill, 
Benny Lautrup, and Steffen B. Petersen 
Training Knowledge-Based Neural Networks to Recognize Genes ....... 
Michiel O. Noordewier, Geoffrey G. TowelL and Jude W. Shavlik 
Neural Network Application to Diagnostics ................. 
Kenneth A. Marko 
Lg Depth Estimation and Ripple Fire Characterization ............ 
John L. Perry and Douglas R. Baumgardt 
A B-P ANN Commodity Trader ...................... 
Joseph E. Collard 
Integrated Segmentation and Recognition of Hand-Printed Numerals ...... 
James D. Keeler, David E. Rumelhart, and Wee-Kheng Leow 
EMPATH: Face, Emotion, and Gender Recognition Using Holons ....... 
Garrison W. Cottrell and Janet Metcalfe 
450 
457 
464 
471 
479 
486 
493 
500 
507 
514 
523 
530 
537 
544 
551 
557 
564 
ix 
SEXNET: A Neural Network Identifies Sex From Human Faces ........ 
B.A. Golomb, D.T. Lawrence, and T.J. Sejnowski 
A Neural Expert System with Automated Extraction of Fuzzy If-Then Rules . . . 
Yoichi Hayashi 
Analog Neural Networks as Decoders ................... 
Ruth Erlanson and Yaser Abu-Mostafa 
Part X Language and Cognition 
Distributed Recursive Structure Processing ................. 
Graldine Legendre, Yoshiro Miyata, and Paul Smolensky 
Translating Locative Prepositions ..................... 
Paul W. Munro and Mary Tabasko 
A Short-Term Memory Architecture for the Learning of Morphophonemic Rules . 
Michael Gasser and Chan-Do Lee 
Exploiting Syllable Structure in a Connectionist Phonology Model ....... 
David S. Touretzky and Deirdre W. Wheeler 
Language Induction by Phase Transition in Dynamical Recognizers ....... 
Jordan B. Pollack 
Discovering Discrete Distributed Representations .............. 
Michael C. Mozer 
Direct Memory Access Using Two Cues ................... 
Janet Wiles, Michael S. Humphreys, John D. Bain, and Simon Dennis 
An Attractor Neural Network Model of Recall and Recognition ........ 
Eytan Ruppin and Yechezkel Yeshurun 
ALCOVE: A Connectionist Model of Human Category Learning ........ 
John K. Kruschke 
Spherical Units as Dynamic Consequential Regions .............. 
Stephen Jos Hanson and Mark A. Gluck 
Connectionist Implementation of a Theory of Generalization 
Roger N. Shepard and Sheila Kannappan 
Part XI Local Basis Functions 
Adaptive Spline Networks ........................ 
Jerome H. Friedman 
Multi-Layer Perceptrons with B-Spline Receptive Field Functions ........ 
Stephen H. Lane, Marshall G. Flax, David A. Handelman, and Jack J. Gelfand 
572 
578 
585 
591 
598 
6O5 
612 
619 
627 
635 
642 
649 
656 
665 
675 
684 
x 
Bumptrees for Efficient Function, Constraint, and Classification Learning .... 
Stephen M. Omohundro 
Basis-Function Trees as a Generalization of Local Variable Selection Methods . . 
Terence D. Sanger 
Generalization Properties of Radial Basis Functions .............. 
Sherif M. Botros and Christopher G. Atkeson 
Learning by Combining Memorization and Gradient Descent ......... 
John C. Platt 
Sequential Adaptation of Radial Basis Function Neural Networks ....... 
V. Kadirkamanathan, M. Niranjan, and F. Fallside 
Oriented Non-Radial Basis Functions for Image Coding and Analysis ...... 
Avijit Saha, Jim Christian, D.S. Tang, and Chuan-Lin Wu 
Computing with Arrays of Bell-Shaped and Sigrnoid Functions ........ 
Pierre Baldi 
Discrete Affine Wavelet Transforms .................... 
Y.C. Pati and P.S. Krishnaprasad 
Extensions of a Theory of Networks for Approximation and Learning ...... 
Federico Girosi, Tomaso Poggio, and Bruno Caprile 
How Receptive Field Parameters Affect Neural Learning ........... 
Bartlett W. Mel and Stephen M. Omohundro 
Part XII Learning Systems 
A Competitive Modular Connectionist Architecture ............. 
Robert A. Jacobs and Michael I. Jordan 
Evaluation of Adaptive Mixtures of Competing Experts ............ 
Steven J. Nowlan and Geoffrey E. Hinton 
A Framework for the Cooperation of Learning Algorithms .......... 
Lon Bottou and Patrick Gallinari 
Connectionist Music Composition Based on Melodic and Stylistic Constraints . . 
Michael C. Mozer and Todd Soukup 
Using Genetic Algorithms to Improve Pattern Classification Performance .... 
Eric I. Chang and Richard P. Lippmann 
Evolution and Learning in Neural Networks ................. 
Ron Keesing and David G. Stork 
Designing Linear Threshold Based Neural Network Pattern Classifiers ..... 
Terrence L. Fine 
693 
700 
707 
714 
721 
728 
735 
743 
750 
757 
767 
774 
781 
789 
797 
804 
811 
xi 
On Stochastic Complexity and Admissible Models for Neural Network Classifiers 818 
Padhraic $myth 
Efficient Design of Boltzmann Machines ................... 825 
Ajay Gupta and Wolfgang Maass 
Note on Learning Rate Schedules for Stochastic Optimization ......... 832 
Christian Darken and John Moody 
Convergence of a Neural Network Classifier ................ 839 
John S. Baras and Anthony LaVigna 
Learning Theory and Experiments with Competitive Networks ........ 846 
Griff L. Bilbro and David E. Van den Bout 
Transforming Neural-Net Output Levels to Probability Distributions ...... 853 
John S. Denker and Yann leCun 
Back Propagation is Sensitive to Initial Conditions . 
John F. Kolen and Jordan B. Pollack 
Closed-Form Inversion of Backpropagation Networks 
Michael L. Rossen 
Part XIII Learning and Generalization 
860 
868 
Generalization by Weight-Elimination with Application to Forecasting ..... 875 
Andreas S. Weigend, David E. Rumelhart, and Bernardo A. Huberman 
The Devil and the Network ........................ 883 
Sanjay Biswas and Santosh S. Venkatesh 
Generalization Dynamics in LMS Trained Linear Networks .......... 890 
Yves Chauvin 
Dynamics of Generalization in Linear Perceptrons .............. 897 
Anders Krogh and John A. Hertz 
Constructing Hidden Units Using Examples and Queries ........... 904 
Eric B. Baum and Kevin J. Lang 
Can Neural Networks do Better Than the Vapnik-Chervonenkis Bounds? .... 911 
David Cohn and Gerald Tesauro 
Second Order Properties of Error Surfaces .................. 918 
Yann Le Cun, Ido Kanter, and Sara A. Solla 
Chaitin-Kolmogorov Complexity and Generalization in Neural Networks .... 925 
Barak A. Pearlmutter and Ronald Rosenfeld 
Asymptotic Slowing Down of the Nearest-Neighbor Classifier ......... 932 
Robert R. Snapp, Demetri Psaltis, and Santosh S. Venkatesh 
xii 
Remarks on Interpolation and Recognition Using Neural Nets ......... 939 
Eduardo D. Sontag 
e-Entropy and the Complexity of Feedforward Neural Networks ........ 946 
Robert C. Williamson 
On The Circuit Complexity of Neural Networks ............... 
V.P. Roychowdhury, A. Orlitsky, K.Y. Siu, and T. Kailath 
953 
Part XIV Performance Comparisons 
Comparison of Three Classification Techniques, CART, C4.5 and Multi-Layer 
Perceptrons ............................... 963 
A.C. Tsoi and R.A. Pearson 
Practical Characteristics of Neural Network and Conventional Pattern Classifiers. 970 
Kenney Ng and Richard P. Lippmann 
Time Trials on Second-Order and Variable-Learning-Rate Algorithms ...... 977 
Richard Rohwer 
Kohonen Networks and Clustering ..................... 984 
Wesley Snyder, Daniel Nissman, David Van den Bout, and Griff Bilbro 
Part XV VLSI 
VLSI Implementations of Learning and Memory Systems ........... 993 
Mark A. Holier 
Compact EEPROM-based Weight Functions ................. 1001 
A. Kramer, C.K. Sin, R. Chu, and P.K. Ko 
An Analog VLSI Splining Network ..................... 1008 
Daniel B. Schwartz and Vijay K. Samalam 
Relaxation Networks for Large Supervised Learning Problems ......... 1015 
Joshua Alspector, Robert B. Allen, Anthony ]ayakumar, Torsten Zeppenfeld, and 
Ronny Meir 
Design and Implementation of a High Speed CMAC Neural Network ...... 1022 
W. Thomas Miller, III, Brian A. Box, Erich C. Whitney, and James M. Glynn 
Back Propagation Implementation ..................... 1028 
Hal McCartor 
Reconfigurable Neural Net Chip with 32K Connections ............ 1032 
H.P. Graf, R. Janow, D. Henderson, and R. Lee 
Simulation of the Neocognitron on a CCD Parallel Processing Architecture . . . 1039 
Michad L. Chuang and Alice M. Chiang 
VLSI Implementation of TInMANN .................... 1046 
Matt Melton, Tan Phan, Doug Reeves, and Dave Van den Bout 
Xlll 
Preface 
Since its inception in 1987, the NIPS conference (short for "Neural Information Processing 
Systems - Natural and Synthetic") has attracted researchers from many disciplines who 
are applying their expertise to problems in the emerging field of neural networks. The 
conference and the following two-day workshop have become a forum for presenting to 
the neural network community the latest research results, and for leading researchers to 
gather and exchange ideas. This volume contains papers summarizing the talks and posters 
presented at the most recent NIPS conference, held in Denver, Colorado, from 26-29 
November 1990. 
The 1990 conference maintained the high level of excitement of its predecessors. Important 
new theoretical and empirical results were presented concerning the capability and gener- 
alization performance of networks. Analyses of biological nervous systems were presented 
along with mathematical models of some of these networks. Many working analog and 
digital VLSI chips were described. These chips demonstrated how rapidly VLSI design 
procedures have advanced in the past few years. New modular learning algorithms were 
also described, along with significant new industrial applications, new theoretical under- 
standing of network system design issues, and significant advances in speech, robotics, 
and machine vision systems. Many papers described the application of dynamic temporal 
networks with recurrent connections. Papers overall demonstrated a better grasp of past 
neural network research and of past theoretical and practical results in other disciplines. 
Multi-disciplinary exchange was promoted this year by grouping papers for presentation 
according to paper topic instead of by discipline. For example, one session on "Visual 
Processing" contained papers on theory, algorithms, VLSI implementation, neurobiology, 
and machine vision applications. Organization by topic instead of by discipline is reflected 
in this volume. Presentation of papers by researchers outsidetheUnited States was promoted 
this year by the addition of foreign liaisons to the conference organizing committee. Our 
current liasons represent Australia, Great Britain, continental Europe, Japan, and South 
America. Presentation of new results by students was promoted this year as in past years. 
Funds for student travel grants were provided by ATR, the Air Force Office of Scientific 
Research, Fujitsu, IBM, NASA, Sanyo, Sharp, Siemens, and the University of Colorado, 
Boulder Optoelectronic Computing Systems Center. 
The NIPS conference continues to be an exciting, successful meeting due to the continued 
efforts of a large group of people. We would first like to thank all the other members of 
the program and organizing committees who helped make this conference possible. In 
particular we would like to thank Judy Terrel for her work throughout the year as the 
conference secretary, Gina Davies and Kathy Hibbard for their work in organizing local 
arrangements and running the conference desk so smoothly, and Alex Waibel for organizing 
the post-conference workshops. Finally, we would like to thank everyone who attended 
and submitted papers and the 85 referees who carefully read and reviewed 20 papers each. 
Richard P. Lippmann 
John E. Moody 
David S. Touretzky 
January, 1991 
NIPS-91 Organizing Committee 
General Chair 
Program Chair 
Publications 
Publicity 
Treasurer 
Local Arrangements 
Workshop Chair 
Workshop Local Arrangements 
IEEE Liaison 
APS Liaison 
Neurobiology Liaison 
Overseas Liaison (Japan) 
Overseas Liaison (Australia, Singapore, India) 
Overseas Liaison (Europe) 
Overseas Liaison (United Kingdom) 
Overseas Liaison (South America) 
Richard P. Lippmann, MIT Lincoln Laboratory 
John E. Moody, Yale 
David Touretzky, CMU 
Stephen Hanson, Siemens Research Center 
Kristina Johnson, University of Colorado, Boulc 
Kathie Hibbard, University of Colorado, Bould, 
Alex Waibel, CMU 
Howard Wachtel, University of Colorado, Boul, 
Edward Posner, Caltech 
Larry Jackel, AT&T Bell Laboratories 
Jim Bower, Caltech 
Mitsuo Kawato, ATR Research Laboratories 
Marwan Jabri, University of Sydney 
Benny Lautrup, Niels Bohr Institute 
John Bridle, RSRE 
Andreas Meier, Simon Bolivar University 
CHAIR 
John E. Moody, Yale 
NIPS-91 Program Committee 
AREA CO-CHAIRS 
Applications: 
Architectures: 
Cognitive Science: 
Implementations: 
Neurobiology: 
Theory: 
Lee Giles, NEC Research Institute 
Yann LeCun, AT&T Bell Labs 
Steve Hanson, Siemens Research Center 
Joshua Alspector, Bellcore 
Terrence Sejnowski, Salk Institute 
Gerry Tesauro, IBM 
REVIEWERS 
Asad Abidi, UCLA 
David Ackley, Bellcore 
Robert Allen, Bellcore 
Luis Almeida, INESC, Portugal 
P. Anandan, Yale 
Dana Anderson, University of Colorado at Boulder 
Christopher Atkeson, MIT 
Pierre Baldi, Caltech/JPL 
Dana Ballard, University of Rochester 
Andrew Barto, University of Massachusetts 
William Bialek, UC Berkeley 
David Bounds, RSRE, England 
Herve Bourlard, Philips Research Lab, Belgium 
James Bower, Caltech 
Thomas Brown, Yale 
Joachim Buhmann, USC 
David Burr, Bellcore 
James Burr, Stanford 
Joseph Change, Yale 
H.H. Chen, University of Maryland 
Jack Cowan, University of Chicago 
John Denker, AT&T Bells Labs 
Georg Dorffner, Austrian Research Institute 
Jeffrey Elman, UCSD 
Terrence Fine, Cornell 
Walter Freeman, UC Berkeley 
Gene Gindi, Yale 
Hans Graf, AT&T Bell Labs 
Kamil Grajski, Ford Aerospace 
Allon Guez, Drexel University 
David Haussler, UC Santa Cruz 
John Hertz, NORDITA, Denmark 
Geoffrey Hinton, University of Toronto 
Mark Holier, Intel 
Nathan Intrator, Brown University 
Larry Jackel, AT&T Bell Labs 
Kristina Johnson, University of Colorado at Boulder 
xvii 
Michael Jordan, MIT 
Steven Judd, Caltech 
Scott Kirkpatrick, IBM 
Christoph Koch, Caltech 
Thomas Landauer, Bellcore 
Benny Lautrup, Neils Bohr Institute, Denmark 
Y.C. Lee, Los Alamos 
Hong Leung, MIT 
Ralph Linsker, IBM 
Richard Lippmann, MIT Lincoln Laboratory 
Richard Mammone, Rutgers 
James Mann, MIT Lincoln Laboratory 
Drew McDermott, Yale 
Bartlett Mel, Woods Hole 
Kenneth Miller, UCSF 
Eric Mjolsness, Yale 
John Moody, Yale 
Ala Murray, University of Edinburgh 
Kumpati Narendra, Yale 
Stephen Omohundro, ICSI, Berkeley 
John Pearson, David Samoff Research Center 
Sandy Pentland, MIT 
Thomas Petsche, Siemens 
Fernando Pineda, JPL/Caltech 
John Platt, Synaptics 
Edward Posner, Caltech 
Michael Roth, John Hopkins 
Jay Sage, MIT Lincoln Labs 
John Schotland, Bellcore 
Eric Schwartz, NYU Medical Center 
Daniel Schwartz, GTE 
Carolo Sequin, UC Berkeley 
Gordon Shepard, Yale 
Josef Skrzypek, UCLA 
Sara Solla, AT&T Bell Laboratories 
David Stork, Stanford 
Richard Sutton, GTE 
Manoel Tenorio, Purdue 
Anil Thakoor, JPL 
David Touretzky, Carnegie Mellon 
David Van Essen, Caltech 
Santosh Venkatesh, University of Pennsylvania 
Hal White, Stanford 
XVlll 
Part I 
Neurobiology 
