Algorithms and Architectures
Regular price
44.750 KD
inc. VAT
Couldn't load pickup availability
Table of contents
- Cover
- Contentsv
- Contributorsxv
- Prefacexix
- Chapter 1. Statistical Theories of Learning in Radial Basis Function Networks1
- I. Introduction1
- II. Learning in Radial Basis Function Networks4
- III. Theoretical Evaluations of Network Performance21
- IV. Fully Adaptive Training„An Exact Analysis40
- V. Summary54
- Appendix55
- References57
- Chapter 2. Synthesis of Three-Layer Threshold Networks61
- I. Introduction62
- II. Preliminaries63
- III. Finding the Hidden Layer64
- IV. Learning an Output Layer73
- V. Examples77
- VI. Discussion84
- VII. Conclusion85
- References86
- Chapter 3. Weight Initialization Techniques87
- I. Introduction87
- II. Feedforward Neural Network Models89
- III. Stepwise Regression for Weight Initialization90
- IV. Initialization of Multilayer Perceptron Networks92
- V. Initial Training for Radial Basis Function Networks98
- VI. Weight Initialization in Speech Recognition Application103
- VII. Conclusion116
- Appendix I: Chessboard 4 X 4116
- Appendix II: Two Spirals117
- Appendix III: GaAs MESFET117
- Appendix IV: Credit Card117
- References118
- Chapter 4. Fast Computation in Hamming and Hopfield Networks123
- I. General Introduction123
- II. Threshold Hamming Networks124
- III. Two-Iteration Optimal Signaling in Hopfield Networks135
- IV. Concluding Remarks152
- References153
- Chapter 5. Multilevel Neurons155
- I. Introduction155
- II. Neural System Analysis157
- III. Neural System Synthesis for Associative Memories167
- IV. Simulations171
- V. Conclusions and Discussions173
- Appendix173
- References178
- Chapter 6. Probabilistic Design181
- I. Introduction181
- II. Unified Framework of Neural Networks182
- III. Probabilistic Design of Layered Neural Networks189
- IV. Probability Competition Neural Networks197
- V. Statistical Techniques for Neural Network Design218
- VI. Conclusion228
- References228
- Chapter 7. Short Time Memory Problems231
- I. Introduction231
- II. Background232
- III. Measuring Neural Responses233
- IV. Hysteresis Model234
- V. Perfect Memory237
- VI. Temporal Precedence Differentiation239
- VII. Study in Spatiotemporal Pattern Recognition241
- VIII. Conclusion245
- Appendix246
- References260
- Chapter 8. Reliability Issue and Quantization Effects in Optical and Electronic Network Implementati261
- I. Introduction261
- II. Hebbian-Type Associative Memories264
- III. Network Analysis Using a Signal-to-Noise Ratio Concept266
- IV. Reliability Effects in Network Implementations268
- V. Comparison of Linear and Quadratic Networks278
- VI. Quantization of Synaptic Interconnections281
- VII. Conclusions288
- References289
- Chapter 9. Finite Constraint Satisfaction293
- I. Constrained Heuristic Search and Neural Networks for Finite Constraint Satisfaction Problems293
- II. Linear Programming and Neural Networks323
- III. Neural Networks and Genetic Algorithms331
- IV. Related Work, Limitations, Further Work, and Conclusions341
- Appendix I. Formal Description of the Shared Resource Allocation Algorithm342
- Appendix II. Formal Description of the Conjunctive Normal Form Satisfiability Algorithm346
- Appendix III. A 3-CNF-SAT Example348
- Appendix IV. Outline of Proof for the Linear Programming Algorithm350
- References359
- Chapter 10. Parallel, Self-Organizing, Hierarchical Neural Network Systems363
- I. Introduction364
- II. Nonlinear Transformations of Input Vectors366
- III. Training, Testing, and Error-Detection Bounds367
- IV. Interpretation of the Error-Detection Bounds371
- V. Comparison between the Parallel, Self-Organizing, Hierarchical Neural Network, the Backpropagatio373
- VI. PNS Modules379
- VII. Parallel Consensual Neural Networks381
- VIII. Parallel, Self-Organizing, Hierarchical Neural Networks with Competitive Learning and Safe Rej385
- IX. Parallel, Self-Organizing, Hierarchical Neural Networks with Continuous Inputs and Outputs392
- X. Recent Applications395
- XI. Conclusions399
- References399
- Chapter 11. Dynamics of Networks of Biological Neurons: Simulation and Experimental Tools401
- I. Introduction402
- II. Modeling Tools403
- III. Arrays of Planar Microtransducers for Electrical Activity Recording of Cultured Neuronal Popula418
- VI. Concluding Remarks421
- References422
- Chapter 12. Estimating the Dimensions of Manifolds Using Delaunay Diagrams425
- I. Delaunay Diagrams of Manifolds425
- II. Estimating the Dimensions of Manifolds435
- III. Conclusions455
- References456
- Index457
This volume is the first diverse and comprehensive treatment of algorithms and architectures for the realization of neural network systems. It presents techniques and diverse methods in numerous areas of this broad subject. The book covers major neural network systems structures for achieving effective systems, and illustrates them with examples.
This volume includes Radial Basis Function networks, the Expand-and-Truncate Learning algorithm for the synthesis of Three-Layer Threshold Networks, weight initialization, fast and efficient variants of Hamming and Hopfield neural networks, discrete time synchronous multilevel neural systems with reduced VLSI demands, probabilistic design techniques, time-based techniques, techniques for reducing physical realization requirements, and applications to finite constraint problems.
A unique and comprehensive reference for a broad array of algorithms and architectures, this book will be of use to practitioners, researchers, and students in industrial, manufacturing, electrical, and mechanical engineering, as well as in computer science and engineering.
Key Features
* Radial Basis Function networks
* The Expand-and-Truncate Learning algorithm for the synthesis of Three-Layer Threshold Networks
* Weight initialization
* Fast and efficient variants of Hamming and Hopfield neural networks
* Discrete time synchronous multilevel neural systems with reduced VLSI demands
* Probabilistic design techniques
* Time-based techniques
* Techniques for reducing physical realization requirements
* Applications to finite constraint problems
* Practical realization methods for Hebbian type associative memory systems
* Parallel self-organizing hierarchical neural network systems
* Dynamics of networks of biological neurons for utilization in computational neuroscience
Practitioners, researchers, and students in industrial, manufacturing, electrical, and mechanical engineering, as well as in computer science and engineering, will find this volume a unique and comprehensive reference to a broad array of algorithms and architectures
This volume includes Radial Basis Function networks, the Expand-and-Truncate Learning algorithm for the synthesis of Three-Layer Threshold Networks, weight initialization, fast and efficient variants of Hamming and Hopfield neural networks, discrete time synchronous multilevel neural systems with reduced VLSI demands, probabilistic design techniques, time-based techniques, techniques for reducing physical realization requirements, and applications to finite constraint problems.
A unique and comprehensive reference for a broad array of algorithms and architectures, this book will be of use to practitioners, researchers, and students in industrial, manufacturing, electrical, and mechanical engineering, as well as in computer science and engineering.
Key Features
* Radial Basis Function networks
* The Expand-and-Truncate Learning algorithm for the synthesis of Three-Layer Threshold Networks
* Weight initialization
* Fast and efficient variants of Hamming and Hopfield neural networks
* Discrete time synchronous multilevel neural systems with reduced VLSI demands
* Probabilistic design techniques
* Time-based techniques
* Techniques for reducing physical realization requirements
* Applications to finite constraint problems
* Practical realization methods for Hebbian type associative memory systems
* Parallel self-organizing hierarchical neural network systems
* Dynamics of networks of biological neurons for utilization in computational neuroscience
Practitioners, researchers, and students in industrial, manufacturing, electrical, and mechanical engineering, as well as in computer science and engineering, will find this volume a unique and comprehensive reference to a broad array of algorithms and architectures