Introduction to the Theory of Neural Computation
Table of Contents
|
By John A. Hertz, Anders S. Krogh, and Richard G. Palmer
Series Foreword
Foreword
Foreword
Preface One: IntroductionInspiration from Neuroscience History The Issues Two: The Hopfield ModelThe Associative Memory Problem The Model Statistical Mechanics of Magnetic Systems Stochastic Networks Capacity of the Stochastic Network Three: Extensions of the Hopfield ModelVariations on the hopfield Model Correlated Patterns Continuous-Valued Units Hardware Implementations Temporal Sequences of Patterns Four: Optimization ProblemsThe Weighted Matching Problem The Travelling Saleman Problem Graph Bipartitioning Optimization Problems in Image Processing Five: Simple PerceptronsFeed-Forward Networks Threshold Units Proof of Convergence of the Perceptron Learning Rule Linear Units Nonlinear Units Stochastic Units Capacity of the Simple Perceptron Six: Multi-Layer NetworksBack-Propagation Variations on Back-Propagation Examples and Applications Performance of Multi-Layer Feed-Forward Networks A Theoretical Framework for Generalization Optimal Network Architectures Seven: Recurrent NetworksBoltzmann Machines Recurrent Back-Propagation Learning Time Sequence Reinforcement Learning Eight: Unsupervised Hebbian LearningUnsupervised Learning One Linear Unit Principal Component Analysis Self-Organizing Feature Extraction Nine: Unsupervised Competitive LearningSimple Competitive Learning Examples and Applications of Competitive Learning Adaptive Resonance Theory Feature Mapping Theory of Feature Mapping The Travelling Salesman Problem Hybrid Learning Schemes Ten: Formal Statistical Mechanics of Neural NetworksThe Hopfield Model Gardner Theory of the Connections Appendix: Statistical MechanicsThe Boltzmann-Gibbs Distribution Free Energy and Entropy Stochastic Dynamics
BibliographySubject Index Author Index Index |
|
