Parallel processing in neural systems and computers

목차 (Table of Contents)

• CONTENTS
• Section 1 : General Introduction
• Neuroinformatics and Cybernetic(Invited Paper) / G. HAUSKE(M$$\ddot u$$nchen, FRG) = 3
• Concerning the Emerging Role of Geometry in Neuroinformatics / R. ECKMILLER(D$$\ddot u$$ssedorf, FRG) = 5
• Philosophical Cencepts of Computational Neuroscience / K. MAINZER(Augsburg, FRG) = 9
• Section 2 : Development and Application of Parallel Computers
• Parallel Processing Using Memory - Based Connectionist Networks / C.-C. CHEN(Brussels, Belgium) = 15
• Accelerating Convergence of Symmetric Neural Networks on a Local Memory Multiprocessor / J. P. COUGHLIN ; R. H. BARAN(Towson, USA) = 19
• Why Do Neural Network Researchers Ignore Stochastic computers? / R. MASSEN(Konstanz, FRG) = 23
• A Programmable Highly Parallel Architecture : Functional Definition and Performance Evaluation / G. MAZARE ; E. PAYAN Z(Grenoble, France) = 27
• A Genetic Algorithm for Massively Parallel Computers / P. SPIESSENS ; B. MANDERICK(Brussels, Belgium) = 31
• A Transputer System for the Recognition of Human Faces by Labeled Graph Matching / R. P. WURTZ ; J. C. VORBR$$\ddot U$$GGEN(Frankfurt, FRG) ; C. v.d. MALSBURG(Los Angeles, USA) = 37
• CORDIC Processor Arrays for Adaptive Least Squares Algorithms / B. YANG ; J. F. B$$\ddot O$$HME(Bochum, FRG) = 43
• Section 3 : New Concepts of Information Processing in the Brain
• Computer Simulation of Neural Networks with Chemical Markers / A. ADAMOPOULOS ; P. ANNINOS(Alexandroupolis, Greece) = 49
• Dynamics of Activity in Neuronal Networks Give Rise to Fast Modulations of Functional Connectivity / K.-H. BOVEN ; A. AERTSEN(T$$\ddot u$$bingen, FRG) = 53
• Using Proprioceptive Information to Control Movement to a Target / J. DEAN(Bielefeld, FRG) = 57
• A New Dynamic Model of REceptive - Field Organization / H. R. DINSE ; J. BEST ; K. KR$$\ddot U$$GER(Bochum, FRG) = 61
• Direct Observation of Neural Assemblies During Neocortical Representational Reorganization / H. R. DINSE(Bochum, FRG) ; G. H. RECANZONE ; M. M. MERZENICHI(San Francisco, USA) = 65
• Large - Scale Simulation of a Self - Organizing Neural Network : Formation of a Somatotopic Map / K. OBERMAYER ; H. RITTER ; K. SCHULTEN(Urbana - Champaign, USA) = 71
• Synchronized Activity in the CA1 - Region of the Rat Hippocampus Reveals Spatio - Temporal CSD - Patterns of Action Potentials / D. PLENZ ; A. AERTSEN(T$$\ddot u$$bingen, FRG) = 75
• Emulation of Biology - Oriented Neural Networks / S. J. PRANGE(Berlin, FRG) = 79
• Are Fractal Dimensions a Good Measure for Neuronal Activity? / H. PREISSEL ; A. AERTSEN(T$$\ddot u$$bingen, FRG) ; G. PALM(D$$\ddot u$$sseldorf, FRG) = 83
• A Dynamic Theory of Coordination of Discrete Movement / G. SCH$$\ddot O$$NER(Bochum, FRG) = 87
• Spike Arrival Times : A Highly Efficient Coding Scheme for Neural Networks / S. J. THORPE(Paris, France) = 91
• Section 4 : Parallel Processing in the Visual System
• Correlated Rhythmic Across Cortical Maps via Synchronization / F. AIPLE ; J. KR$$\ddot U$$GER(Freiburg, FRG) = 97
• Feature Linking Across Cortical Maps via Synchronization / R. ECKHORN ; H. J. REITBOECK ; P. DICKE ; M. ARNDT ; W. KRUSE(Marburg, FRG) = 101
• Synchronization of Oscillatory Responses : A Mechanism for Stimulus - Dependent Assembly Formation in Cat Visual Cortex / A. K. ENGEL ; P. K$$\ddot O$$NIG ; C. M. GRAY ; W. SINGER(Frankfurt, FRG) = 105
• Parallel Distributed Processing of Configural Features of Moving Objects in the Toad's Visual System / J.-P. EWERT ; W. W. SCHWIPPERT ; T. W. BENEKE(Kassel, FRG) = 109
• The Thalamo - Cortical Feedback - System : A Nonlinear Network - Model for Binocular Interaction / D. HEINKE ; F. GIANNAKOPOULOS ; J. BEST(Bochum, FRG) = 113
• Segregation of Oscillatory Responses by Conflicting Stimuli - Desynchronizing Connections in Neural Oscillator Layers / P. K$$\ddot O$$NIG ; T. B. SCHILLEN(Frankfurt, FRG) = 117
• Segregation of "Meaning" and "Importance" of Neuronal Messages / J. KR$$\ddot U$$GER ; J. D. BECKER(Freiburg, FRG) = 121
• Retinal Sampling Grids and Space - Variant Image Compression / H. A. MALLOT ; G.-J. GIEFING(Bochum, FRG) = 125
• Towards a Network Theory of Cortical Areas / H. A. MALLOT ; W. VON SEELEN(Bochum, FRG) = 129
• Processing of Figure and Background Motion in the Visual System of the Fly(Invited Paper) / W. REIGHARDT(T$$\ddot u$$bingen, FRG) = 133
• Simulation of Delayed Oscillators with the MENS General Purposoe Modeling Environment for Network Layers / T. B. SCHILLEN(Frankfurt, FRG) = 135
• Coherency Detection by Coupled Oscillatory Responses - Synchronizing Connections in Neural Oscillator Layers / T. B. SCHUSTER ; P. K$$\ddot O$$NIG(Frankfurt) = 139
• A Model for Neuronal Oscillations in the Visual Cortex / H. G. SCHUSTER ; P. WAGNER(Kiel, FRG) = 143
• Temporal Dynamics in Neuronal Microcircuitry / F. W$$\ddot O$$RG$$\ddot O$$TTER ; D. M. KAMMEN ; B. BRANDT(Pasadena, USA) = 147
• Section 5 : Self - Organization and Learning in Neural Networks
• Properties of an Adaptive Perceptron Algorithm / J. K. ANLAUF ; M. BIEHL(Giessen, FRG) = 153
• Learning Symbol Processing with Recurrent Networks / R. GOEBEL(Marburg, FRG) = 157
• Training Strategies for Probabilistic RAMs / D. GORSE ; J. G. TAYLOR(London, UK) = 161
• Learning in Neural Nets by Genetic Algorithms / J. HEISTERMANN(M$$\ddot u$$nchen, FRG) = 165
• A Neural Network which Adapts its Structure to a Given Set of Patterns / S. JOCKUSCH(G$$\ddot o$$ttingen, RFG) = 169
• Efficient Learning Alorithms for Single - Layered Nerual Networks / N. B. KARAYIANNIS ; A, N, VENETSANOPOLUS(Toronto, Canada) = 173
• Internal Representations and Accociative Memory(Invited Paper) / T. KOHONEN(Espoo, Finland) = 177
• Hebbian Learning of Principal Components / A. KROGH ; J. A. HERTZ(Copenhagen, Denmark) = 183
• The Anti - Hebb Rule Derived from Information Theory / H. K$$\ddot U$$HNEL ; P. TAVAN(Munchen, FRG) = 187
• Dynamical Learning in Networks with Sparse Connectivity / K. E. K$$\ddot U$$RTEN(Bochum, FRG) = 191
• Self - Organization versus Programming in Massively Parallel Systems : A Case Study / B. MANDERICK ; F. MOYSON(Brussels, Belgium) = 195
• Top - Down Learning in Modular Feed - Forward Networks / A. OSSEN(Berlin, FRG) = 201
• Genetic Generation of Backpropagation Trained Neural Networks / W. SCHIFFMANN ; K. MECKLENBURG(Koblenz, FRG) = 205
• Temporal - Difference - Driven Learning in Recurrent Networks / J. SCHMIDUHUBER(M$$\ddot u$$nchen, FRG) = 209
• Associative Memory versus Linear Filter : Distinct Operational Modes of Stable Artificial Neural Systems / B. SCH$$\ddot U$$RMANN ; I. LEUTH$$\ddot A$$USSER ; J. HOLLATZ(M$$\ddot u$$chen, FRG) = 213
• Compectitive Sequence Learning / C. WINDHEUSER ; J. KINDERMANN(St. Augustink, FRG) = 217
• Section 6 : Evaluation of Artificial Neural Networks
• Ideal Neurons for Neura Computers / I. ALEKSANDER(London, UK) = 225
• On the Biologically Motivated Derivation of Kohonen's Self - Organizing Feature Maps / R. ACKER ; A. KURZ(Darmastadt, FRG) = 229
• All - or - None Connections of Variable Weights? / A. DE CALLATAY(La Hulpe, Belgium) = 233
• Learning to Predict the Consequences of One's Own Actions / F. CECCONI ; D. PARISI(Rome, Italy) = 237
• Various Roles for Inhibitory Interneurons in the Formation of Associative Memory / I. E. DAMMASH ; J. R. WOLFF(G$$\ddot o$$ttingen, FRG) = 241
• Soluble Low - Activity Networks with Nonlinear Synapses / H. ENGLISCH(Leipzig, GDR) ; J. L. HEMMEN(M$$\ddot u$$nchen, FRG) = 245
• Implementation of Fuzzy Production Systems with Neural Networks / W. EPPLER(Karlsruhe, FRG) = 249
• A Neural Network for Job Sequencing / L. FANG ; W. H. WILSON(Bedford Park, Australia) ; T. LI(Clayton, Australisa) = 253
• Increasing th Storage Capacity and Shaping the Basins of Attraction of Neural Nets / S. GEVA ; J. SITTE(Brisbane, Australia) = 257
• Drift and Diffusion in Backpropagation Networks / G. RADONS ; H. G. SCHUSTER ; D. WERNER(Kiel, FRG) = 261
• Solving Constraint Problems Using Feedback Neural Networks / H. N. SCHALLER(M$$\ddot u$$nchen, FRG) = 265
• The Exact Evaluation of Memory Capacity for New Stochastic Versions of the Hopfield Net / P. WHITTLE(Cambridge, UK) = 269
• Sparse Coding : The Link Between Low Level Vision and Accociative Memory / C. ZETZSCHE(M$$\ddot u$$nchen, FRG) = 273
• The Convergence of Parallel Boltzmann Machines / P. J. ZWIETERING ; E. H. L. AARTS(Eindhoven, The Netherlands) = 277
• Section 7 : Hardware and Software Simulators for Neural Networks
• A Transputer Based General Simulator for Connectionist Models / H. P. ERNST ; B. MOKRY ; Z. SCHRETER(Z$$\ddot u$$rich, Switzerland) = 283
• Neural Pascal : A Language for Neural Network Programming / H. P. GUMM(New Paltz, USA) ; F. B. HERGERT(M$$\ddot u$$nchen, FRG) = 287
• Modeling for Neuronal Systems on Transputer Networks / M. MIGLIORE ; G. F. AYALA ; S. L. FORNILI(Palermo, Italy) = 291
• A. Development Tool for Neural Networks Simulations on Transputers / M. MIKSA(Marburg, FRG) = 295
• A Software Environment for Flexible and Rapid Prototyping of Neural Network Applications / J. A. G. NUHUIS ; P. E. DE HAAN ; L. SPAANENBURG ; F. WARKOWSKI(Stuttgart, FRG) = 299
• Systolic Simulation of Multilayer, Feedforward Neural Networks / N. PETKOV(Erlangen, FRG) = 303
• Artificial Neural Network for Real - Time Task Allocation in Fault - Tolerant, Distributed Processing System / P. W. PROTZEL(Hampton, USA) = 307
• Dustributed Processing Hardware for Realization of Artificial Neural Networks / P. RICHERT ; G. HESS ; B. HOSTICKA ; M. KEPSER ; M. SCHWARZ(Duisburg, FRG) = 311
• Multiprocessor Simulation of a Self - Organizing Neural Network on NERV / R. STOTZKA ; R. HAUSER ; R. M$$\ddot A$$NNER(Heidelberg, FRG) = 315
• IC³: A Neural ASIC for Real - Time Prototyping / F. WORKOWSKI ; L. SPAANEBURG ; J. A. G. NIJHUS(Stuttgart, FRG) = 319
• New Concepts for Information Processing in Connectionistic Systems / T. WASCHULZIK ; H. GEIGER ; M. ARNOLDI ; D. B$$\ddot O$$LLER ; A. NISCHWITZ ; W. BRAUER(M$$\ddot u$$nchen, FRG) = 323
• Section 8 : Neural Networks for Visual Pattern Recognition
• Recogniton of New Spatio - Temporal Patterns by Adptive Juction / Y. AJIOKA ; Y. ANZAI ; H. AISO(Yokohama, Japan) = 331
• Hierarchical Spin Model for Stereo Interpretation Using Phase Sensitive Detectors / R. DIVKO ; K. SCHULTEN(M$$\ddot u$$nchen, FRG) = 335
• A Shift Invariant Network Utilizing the GRT / M. FANG ; G. H$$\ddot A$$USLER(Erlangen, FRG) = 339
• Applying the ART1 Architecture to a Pattern Recognition Task / E. C. D. B. C. FILHO ; D. L. BISSET(Kent, UK) = 343
• Neural Network Models for Visual Pattern Recognitin(Invited Paper) / K. FUKUSHIMA(Osaka, Japan) = 351
• $$\sum \prod $$ - Networks for Motion and Invariant Form Analyses / H. GL$$\ddot U$$NDER(M$$\ddot u$$nchen, FRG) = 357
• Self Organization of a Network Linking Features by Synchronization / G. HARTMANN ; S. DR$$\ddot U$$E(Paderborn, FRG) = 361
• A Self - Organizing Network for Complete Feature Extraction / J. RUBNER ; K. SCHULTEN ; P. TAVAN(M$$\ddot u$$nchen, FRG) = 365
• A Self - Organising Neural Net for Depth Movement Analysis / F. SEYTTER(Edinburgh, UK) = 369
• Section 9 : Neural Networks for Auditory Pattern Recognition
• Sequence Analysis in Feedback Multilayer Perceptrons / H.-U. BAUER ; T. GEISEL(Frankfurt, FRG) = 375
• A Neural Net for Recognition and Storing of Spoken Words / H. BEHME(G$${\ddot o}$$ttingen, FRG) = 379
• Adaptive Resonance Theory : Neural Network Architectures for Self - Organizing Pattern Recognition(Invited Paper) / G. A. CARPENTER ; S. GROSSBERG(Boston, USA) = 383
• Word Recognition with a Recurrent Neural Network / F. KOWALEWSKI ; H. W. STRUBE(G$${\ddot o}$$ttingen, FRG) = 391
• A Feedback Network for Temporal Pattern Recognition / T. B. LUDERMIR(London, UK) = 395
• Spectral Processing of Harmonic Complex Tones and Pitch by PDP Networks / R. W. W. TOMLINSON ; W. TREURENIET(Darmstadt, FRG and Ottwa, Canada) = 399
• Section 10 : Neural Networks for Motor Control
• Resistive Network Approach for Obstacle Avoidance in Trajectory Planning / J. BECKMANN(D$${\ddot u}$$sseldorf, FRG) = 405
• A Simple Network Controlling the Movement of a Three Joint Planar Manipulator / H. CRUSE ; M. BR$${\ddot U}$$WER(Bielefeld, FRG) = 409
• Optimal Solution of Underdetermined Linear Matrix Equations in Neural Networks / W. J. DAUNICHT ; H. WERNTGES(D$${\ddot u}$$sseldorf, FRG) = 413
• Inverse Kinematics with Obstacle Avoidance Implemented as a DEFAnet / W. J. DAUNICHT ; M. LADES ; H. WERNTGES ; R. ECKMILLER(D$${\ddot u}$$sseldorf, FRG) = 417
• Parallel Neural - Net Path - Planner in Hypercubes and Transputers / A. W. HO ; G. C. FOX(Pasadena, USA) = 421
• A Control Concept for a Multifingered Robot Gripper Based on Neuron - Like Associative Memories / M. HORMEL(Darmstadt, FRG) = 427
• Learning of Visuomotor - Coordination of a Robot Arm with Redundant Degrees of Freedom / T. MARTINETZ ; H. RITTER ; K. SCHULTEN(Urbana - Champaign, USA) = 431
• Delta Rule - Based Neural Networks May Be Applicable to a Control Task that Does Not Provide a "Teacher" / H. W. WERNTGES(D$${\ddot u}$$sseldorf, FRG) = 435
• Section 11 : Selected Applications for Neural Networks
• Parallel Algorithms for Channel Assignment in Cellular Mobile Radio Systems : The Neural Network Approach / M. DUQUE ANT$${O'}$$N ; D. KUNZ(Hamburg, FRG) = 441
• Storing Cycles in Analog Neural Networks / T. GENCIC ; M. LAPPE ; G. DANGELMAYR ; W. C$${\ddot U}$$TTINGER(T$${\ddot u}$$bingen, FRG) = 445
• Recognition of Pattterns and Movement Patterns by a Synergetic Computet(Invited Paper) / H. HAKEN(Stuttgart, FRG) = 451
• Pattern Recognition in Damaged Hopfield Networks / E. KOSCIELNY - BUNDE(Hamburg, FRG) = 459
• Section 12 : Parallel Processing in Artificial Intelligence
• Parallel Process Interfaces to Knowledge Systems(Invited Paper : Cremers) / K. H. BECKS ; W. BURGARD ; A. B. CREMERS ; A. HEMKER ; A. ULTSCH(Dortmund and Wuppertal, FRG) = 465
• Motion Detection by Correlation and Voting / S. BOHRER ; H. H. B$${\ddot U}$$LTHOFF ; H. A. MALLOT(Bochum, FRG and Providence, USA) = 471
• The Development of "Symbolic Behaviour" in Natural and Artificial Neural Networks / G. D. A. BROWN ; M. OAKSFORD(Bangor, UK) = 475
• Apparent Motion and Other Mysteries(Invited Paper) / J. A. FELDMAN(Berkeley, USA) = 479
• Learning to Product Disriminative Object Descriptions : On the Representation of Rules in a PDP - Net / R. MANGOLD - ALLWINN(Mannheim, FRG) = 487
• Toward a Computational Architecture for Monocular Preattentive Segmentation / H. NEUMANN ; H. S. STIEHL(Hamburg, FRG) = 491
• Modelling Attention in a Connectionist Speech Production Model / U. SCHADE ; H. J. EIKMEYER(Bielefeld, FRG) = 495
• Learning of Control Knowledge for Symbolic Proofs with Backpropagation Networks / A. ULTSCH ; R. HANNUSCHKA ; U. HARTMANN, WEBER(Dortmund, FRG) = 499
• Intelligent Dimensional Data - Reduction by a Topological Map / L. VERCAUTEREN ; R. A. VINGERHOEDS ; L. BOULLART(Ghent, Belgium) = 503
• Parallel Associative Precesses in Informatiion Retrieval / M. WETTER ; R. RAPP(Paderborn, FRG) = 509
• Dustributed Parallel Cooperative Problem - Solving with Voting and Election System of Neural Learning Networks / B. T. ZHANG ; G. VEENKER(Bonn, FRG) = 513
• Section 13 : Optical and Molecular Computing
• Novel Logic and Architectures for Molecular Computing(Invited Paper) / J. R. BARKER(Glasgow, UK) = 519
• New Laser Techniques for Quasi - Molecular Storage(Invited Paper) / D. HAARER(Bayreuth, FRG) = 525
• Chaos, Cooperation, and Associative Memory in Nonlinear Pictorial Feedback System / G. H$${\ddot A}$$USLER(Erlangen, FRG) = 533
• Optically Controlled Multistable Quantum System : Possible Realization of a Molecular Computer? / G. MAHLER ; W. G. TEICH(Stuttgart, FRG) = 539
• Learning in Optical Neural Networks(Invited Paper) / D. PA\SALTIS ; D. BRADY ; K. HSU(Pasadena, USA) = 543
• References from All Contributions = 549
• List of Contributors = 599
• Author Index = 611
• Subject Index = 623