Ebook: Neural Networks: An Introduction

Neural Networks The concepts of neural-network models and techniques of parallel distributed processing are comprehensively presented in a three-step approach: - After a brief overview of the neural structure of the brain and the history of neural-network modeling, the reader is introduced to associative memory, preceptrons, feature-sensitive networks, learning strategies, and practical applications. - The second part covers more advanced subjects such as the statistical physics of spin glasses, the mean-field theory of the Hopfield model, and the "space of interactions" approach to the storage capacity of neural networks. - In the self-contained final part, seven programs that provide practical demonstrations of neural-network models and their learning strategies are discussed. The software is included on a 3 1/2-inch MS-DOS diskette. The source code can be modified using Borland's TURBO-C 2.0 compiler, the Microsoft C compiler (5.0), or compatible compilers.

---

Neural Networks The concepts of neural-network models and techniques of parallel distributed processing are comprehensively presented in a three-step approach: - After a brief overview of the neural structure of the brain and the history of neural-network modeling, the reader is introduced to associative memory, preceptrons, feature-sensitive networks, learning strategies, and practical applications. - The second part covers more advanced subjects such as the statistical physics of spin glasses, the mean-field theory of the Hopfield model, and the "space of interactions" approach to the storage capacity of neural networks. - In the self-contained final part, seven programs that provide practical demonstrations of neural-network models and their learning strategies are discussed. The software is included on a 3 1/2-inch MS-DOS diskette. The source code can be modified using Borland's TURBO-C 2.0 compiler, the Microsoft C compiler (5.0), or compatible compilers.

---

Neural Networks The concepts of neural-network models and techniques of parallel distributed processing are comprehensively presented in a three-step approach: - After a brief overview of the neural structure of the brain and the history of neural-network modeling, the reader is introduced to associative memory, preceptrons, feature-sensitive networks, learning strategies, and practical applications. - The second part covers more advanced subjects such as the statistical physics of spin glasses, the mean-field theory of the Hopfield model, and the "space of interactions" approach to the storage capacity of neural networks. - In the self-contained final part, seven programs that provide practical demonstrations of neural-network models and their learning strategies are discussed. The software is included on a 3 1/2-inch MS-DOS diskette. The source code can be modified using Borland's TURBO-C 2.0 compiler, the Microsoft C compiler (5.0), or compatible compilers.
Content:
Front Matter....Pages i-xv
Front Matter....Pages 1-1
The Structure of the Central Nervous System....Pages 3-12
Neural Networks Introduced....Pages 13-23
Associative Memory....Pages 24-37
Stochastic Neurons....Pages 38-45
Cybernetic Networks....Pages 46-51
Multilayered Perceptrons....Pages 52-62
Applications....Pages 63-71
More Applications of Neural Networks....Pages 72-92
Network Architecture and Generalization....Pages 93-107
Associative Memory: Advanced Learning Strategies....Pages 108-125
Combinatorial Optimization....Pages 126-134
VLSI and Neural Networks....Pages 135-143
Symmetrical Networks with Hidden Neurons....Pages 144-150
Coupled Neural Networks....Pages 151-161
Unsupervised Learning....Pages 162-173
Evolutionary Algorithms for Learning....Pages 174-187
Front Matter....Pages 189-189
Statistical Physics and Spin Glasses....Pages 191-200
The Hopfield Network for p/N ? 0....Pages 201-208
The Space of Interactions in Neural Networks....Pages 209-230
Front Matter....Pages 231-245
Numerical Demonstrations....Pages 247-247
ASSO: Associative Memory....Pages 249-252
ASSCOUNT: Associative Memory for Time Sequences....Pages 253-263
PERBOOL: Learning Boolean Functions with Back-Prop....Pages 264-267
PERFUNC: Learning Continuous Functions with Back-Prop....Pages 268-274
Solution of the Traveling-Salesman Problem....Pages 275-278
KOHOMAP: The Kohonen Self-organizing Map....Pages 279-290
BTT: Back-Propagation Through Time....Pages 291-295
NEUROGEN: Using Genetic Algorithms to Train Networks....Pages 296-302
Back Matter....Pages 303-306
....Pages 307-331