Proceedings of the 1993 Connectionist Models Summer School

by Mozer; Michael C.

• ISBN13: 9780805815900
• ISBN10: 0805815902
• Edition: 1st
• Format: Paperback
• Copyright: 1993-11-01
• Publisher: Psychology Pres

Summary

The result of the 1993 Connectionist Models Summer School, the papers in this volume exemplify the tremendous breadth and depth of research underway in the field of neural networks. Although the slant of the summer school has always leaned toward cognitive science and artificial intelligence, the diverse scientific backgrounds and research interests of accepted students and invited faculty reflect the broad spectrum of areas contributing to neural networks, including artificial intelligence, cognitive science, computer science, engineering, mathematics, neuroscience, and physics. Providing an accurate picture of the state of the art in this fast-moving field, the proceedings of this intense two-week program of lectures, workshops, and informal discussions contains timely and high-quality work by the best and the brightest in the neural networks field.

Table of Contents

Foreword	p. vii
Participants in the 1993 Connectionist Models Summer School	p. ix
Neuroscience	p. 1
Sigma Vs Pi Properties of Spiking Neurons	p. 3
Sigma Vs Pi Properties of Spiking Neurons	p. 3
Towards a Computational Theory of Rat Navigation	p. 11
Conclusion	p. 18
References	p. 18
Evaluating Connectionist Models in Psychology and Neuroscience	p. 20
Acknowledgments	p. 25
References	p. 26
Vision	p. 29
Self-Organizing Feature Maps with Lateral Connections: Modeling Ocular Dominance	p. 31
References	p. 37
References	p. 37
Joint Solution of Low, Intermediate and High-Level Vision Tasks by Global Optimization: Application to Computer Vision at Low Snr	p. 39
Joint Solution of Low, Intermediate and High-Level Vision Tasks by Global Optimization: Application to Computer Vision at Low Snr	p. 39
Learning Global Spatial Structures from Local Associations	p. 48
Learning Global Spatial Structures from Local Associations	p. 48
References	p. 54
Acknowledgments	p. 54
Cognitive Modeling	p. 55
Connectionist Model of Auditory Morse Code Perception	p. 57
Connectionist Model of Auditory Morse Code Perception	p. 57
Acknowledgments	p. 64
References	p. 64
Competitive Neural Network Model for the Process of Recurrent Choice	p. 65
Competitive Neural Network Model for the Process of Recurrent Choice	p. 65
Acknowledgements	p. 72
Acknowledgements	p. 72
A Neural Network Simulation of Numerical Verbal-To-Arabic Transcoding	p. 73
Conclusions	p. 78
Acknowledgments	p. 79
References	p. 79
Combining Models of Single-Digit Arithmetic and Magnitude Comparison	p. 81
Acknowledgements	p. 86
References	p. 86
Neural Network Models as Tools for Understanding High-Level Cognition: Developing Paradigms for Cognitive Interpretation of Neural Network Models	p. 87
Neural Network Models as Tools for Understanding High-Level Cognition: Developing Paradigms for Cognitive Interpretation of Neural Network Models	p. 87
Conclusions	p. 92
Acknowledgments	p. 92
References	p. 93
Language	p. 95
Modeling Language as Sensorimotor Coordination	p. 97
Structure and Content in Word Production: Why It's Hard to Say Dlorm	p. 105
Acknowledgments	p. 111
Acknowledgments	p. 111
Investigating Phonological Representations: a Modeling Agenda	p. 113
Acknowledgments	p. 120
Part-Of-Speech Tagging Using a Variable Context Markov Model	p. 122
Part-Of-Speech Tagging Using a Variable Context Markov Model	p. 122
References	p. 129
Acknowledgment	p. 129
Quantitative Predictions from a Constraint-Based Theory of Syntactic Ambiguity Resolution	p. 130
Optimality Semantics	p. 138
Optimality Semantics	p. 138
Symbolic Computation and Rules	p. 147
What's in a Rule? the Past Tense by Some Other Name Might Be Called a Connectionist Net	p. 149
What's in a Rule? the Past Tense by Some Other Name Might Be Called a Connectionist Net	p. 149
References	p. 156
Acknowledgements	p. 156
On the Proper Treatment of Symbolism a Lesson from Linguistics	p. 157
Conclusion	p. 160
Acknowledgments	p. 160
References	p. 161
Structure Sensitivity in Connectionist Models	p. 162
Conclusion	p. 168
Acknowledgments	p. 168
References	p. 169
Looking for Structured Representations in Recurrent Networks	p. 170
Looking for Structured Representations in Recurrent Networks	p. 170
Conclusion	p. 176
Acknowledgements	p. 176
References	p. 177
Back Propagation with Understandable Results	p. 178
Back Propagation with Understandable Results	p. 178
Introduction	p. 178
Conclusion	p. 183
References	p. 183
Understanding Neural Networks Via Rule Extraction and Pruning	p. 184
Understanding Neural Networks Via Rule Extraction and Pruning	p. 184
Acknowledgements	p. 190
Conclusions	p. 190
References	p. 191
Rule Learning and Extraction with Self-Organizing Neural Networks	p. 192
Concluding Remarks	p. 199
Acknowledgements	p. 199
Acknowledgements	p. 199
Recurrent Networks and Temporal Pattern Processing	p. 201
Recurrent Networks: State Machines or Iterated Function Systems?	p. 203
Recurrent Networks: State Machines or Iterated Function Systems?	p. 203
Acknowledgments	p. 210
References	p. 210
On the Treatment of Time in Recurrent Neural Networks	p. 211
Acknowledgements	p. 217
References	p. 218
Finding Metrical Structure in Time	p. 219
Conclusions	p. 225
Acknowledgments	p. 226
Representations of Tonal Music: a Case Study in the Development of Temporal Relationships	p. 228
Representations of Tonal Music: a Case Study in the Development of Temporal Relationships	p. 228
Conclusions	p. 233
Acknowledgments	p. 234
Applications of Radial Basis Function Fitting to the Analysis of Dynamical Systems	p. 236
Acknowledgments	p. 243
References	p. 243
Event Prediction: Faster Learning in a Layered Hebbian Network with Memory	p. 245
Event Prediction: Faster Learning in a Layered Hebbian Network with Memory	p. 245
Acknowledgments	p. 250
References	p. 250
Control	p. 253
Issues in Using Function Approximation for Reinforcement Learning	p. 255
Approximating Q-Values with Basis Function Representations	p. 264
Approximating Q-Values with Basis Function Representations	p. 264
References	p. 271
References	p. 271
Efficient Learning of Multiple Degree-Of-Freedom Control Problems with Quasi-Independent Q-Agents	p. 272
Efficient Learning of Multiple Degree-Of-Freedom Control Problems with Quasi-Independent Q-Agents	p. 272
References	p. 279
Neural Adaptive Control of Systems with Drifting Parameters	p. 280
Learning Algorithms and Architectures	p. 289
Temporally Local Unsupervised Learning: The Maxin Algorithm for Maximizing Input Information	p. 291
Minimizing Disagreement for Self-Supervised Classification	p. 300
Acknowledgements	p. 307
References	p. 307
Comparison of Two Unsupervised Neural Network Models for Redundancy Reduction	p. 308
Conclusion	p. 314
References	p. 315
Acknowledgments	p. 315
Solving Inverse Problems Using an Em Approach to Density Estimation	p. 316
Estimating A-Posteriori Probabilities Using Stochastic Network Models	p. 324
Estimating A-Posteriori Probabilities Using Stochastic Network Models	p. 324
References	p. 331
References	p. 331
Learning Theory	p. 333
On Overfitting and the Effective Number of Hidden Units	p. 335
Increase of Apparent Complexity is Due to Decrease of Training Set Error	p. 343
Acknowledgments	p. 348
References	p. 348
Momentum and Optimal Stochastic Search	p. 351
Momentum and Optimal Stochastic Search	p. 351
References	p. 356
Acknowledgments	p. 356
Scheme to Improve the Generalization Error	p. 358
Scheme to Improve the Generalization Error	p. 358
Acknowledgements	p. 363
References	p. 363
General Averaging Results for Convex Optimization	p. 364
General Averaging Results for Convex Optimization	p. 364
Acknowledgements	p. 369
Summary	p. 369
References	p. 370
Multitask Connectionist Learning	p. 372
Acknowledgements	p. 378
Summary	p. 378
References	p. 379
Estimating Learning Performance Using Hints	p. 380
Conclusion	p. 385
References	p. 386
Acknowledgement	p. 386
Simulation Tools	p. 387
A Simulator for Asynchronous Hopfield Models	p. 389
An Object-Oriented Dataflow Approach for Better Designs of Neural Net Architectures	p. 397
Acknowledgments	p. 404
References	p. 404
Index	p. 405
Table of Contents provided by Publisher. All Rights Reserved.