Implementation Techniques
Regular price
43.750 KD
inc. VAT
Couldn't load pickup availability
Table of contents
- Cover
- Title Pageiii
- Copyright Pageiv
- Contentsv
- Contributorsxiii
- Prefacexv
- Chapter 1. Recurrent Neural Networks: Identification and Other System Theoretic Properties1
- I. Introduction1
- II. Recurrent Neural Networks4
- III. Mixed Networks17
- IV. Some Open Problems46
- References47
- Chapter 2. Boltzmann Machines: Statistical Associations and Algorithms for Training51
- I. Introduction51
- II. Relationship with Markov Chain Monte Carlo Methods53
- III. Deterministic Origins of Boltzmann Machines55
- IV. Hidden Units56
- V. Training a Boltzmann Machine57
- VI. An Example with No Hidden Unit67
- VII. Examples with Hidden Units73
- VIII. Variations on the Basic Boltzmann Machine81
- IX. The Future Prospects for Boltzmann Machines86
- References87
- Chapter 3. Constructive Learning Techniques for Designing Neural Network Systems91
- I. Introduction91
- II. Classification95
- III. Regression Problems111
- IV. Constructing Modular Architectures124
- V. Reducing Network Complexity129
- VI. Conclusion134
- VII. Appendix: Algorithms for Single-Node Learning135
- References139
- Chapter 4. Modular Neural Networks147
- I. Introduction147
- II. Why Modular Networks?149
- III. Modular Network Architectures152
- IV. Input Decomposition152
- V. Output Decomposition154
- VI. Hierarchical Decomposition157
- VII. Combining Outputs of Expert Modules159
- VIII, Adaptive Modular Networks167
- IX. Conclusions177
- References178
- Chapter 5. Associative Memories183
- I. Introduction183
- II. Point Attractor Associative Memories192
- III. Continuous PAAM: Competitive Associative Memories213
- IV. Discrete PAAM: Asymmetric Hopfield-Type Networks232
- V. Summary and Concluding Remarks References250
- Chapter 6. A Logical Basis for Neural Network Design259
- I. Motivation259
- II. Overview262
- III. Logic, Probability, and Bearing273
- IV. Principle of Maximized Bearing and ILU Architecture278
- V. Optimized Transmission293
- VI. Optimized Transduction296
- VII. ILU Computational Structure300
- VIII. ILU Testing302
- IX. Summary305
- Appendix: Significant Marginal and Conditional ILU Distributions305
- References307
- Chapter 7. Neural Networks Applied to Data Analysis309
- I. Introduction309
- II. Data Complexity311
- III. Data Separability316
- IV. Classifier Selection327
- V. Classifier Nonlinearity341
- VI. Classifier Stability354
- VII. Conclusions and Discussion366
- References367
- Chapter 8. Multimode Single-Neuron Arithmetics371
- I. Introduction371
- II. Defining Neuronal Arithmetics374
- III. Phase Space of Neuronal Arithmetics378
- IV. Multimode Neuronal Arithmetic Unit387
- V. Toward a Computing Neuron392
- VI. Summary395
- References395
- Index397
This volume covers practical and effective implementation techniques, including recurrent methods, Boltzmann machines, constructive learning with methods for the reduction of complexity in neural network systems, modular systems, associative memory, neural network design based on the concept of the Inductive Logic Unit, and a comprehensive treatment of implementations in the area of data classification. Numerous examples enhance the text. Practitioners, researchers,and students in engineering and computer science will find Implementation Techniques a comprehensive and powerful reference.
Key Features
* Recurrent methods
* Boltzmann machines
* Constructive learning with methods for the reduction of complexity in neural network systems
* Modular systems
* Associative memory
* Neural network design based on the concept of the Inductive Logic Unit
* Data classification
* Integrated neuron model systems that function as programmable rational approximators
With numerous examples to enhance the text, practitioners, researchers, and students in engineering and computer science will find Implementation Techniques a uniquely comprehensive and powerful reference source
Key Features
* Recurrent methods
* Boltzmann machines
* Constructive learning with methods for the reduction of complexity in neural network systems
* Modular systems
* Associative memory
* Neural network design based on the concept of the Inductive Logic Unit
* Data classification
* Integrated neuron model systems that function as programmable rational approximators
With numerous examples to enhance the text, practitioners, researchers, and students in engineering and computer science will find Implementation Techniques a uniquely comprehensive and powerful reference source