| Description |
1 online resource (257 pages) |
| Contents |
Cover; Half Title; Title Page; Copyright Page; Dedication; Table of Contents; Preface and Acknowledgements; Part I CALM: Categorizing And Learning Module; 1 Introduction; 1.1 The importance of learning; 1.2 Some problems with learning neural networks; 1.3 Structural constraints; 1.4 Functional constraints: attention and learning; 1.5 Implementation of the constraints; 2 Description of CALM; 2.1 Structure of CALM; 2.2 Functioning of CALM; 3 Simulation studies of performance and self-organization in CALM; 3.1 Simulation with CALM; 3.2 Convergence time and discrimination time |
|
3.3 Discrimination and clustering3.4 Illustration: learning the EXOR; 3.5 Topological self-organization in CALM modules; Part II Application; 4 Psychological models; 4.1 Learning the word-superiority effect for letter recognition; 4.2 Modelling human memory; 5 Pattern recognition as a practical application; 5.1 Approaches to pattern recognition; 5.2 Sources o f pattern variability; 5.3 A small network that learns handwritten numerals; 6 Genetic algorithms: modularity, learning and network design; 6.1 A brief introduction to genetic algorithms; 6.2 Modules as partial solutions |
|
6.3 The interaction of evolution and learning6.4 Genetic algorithms and neural networks; 6.5 Designing modular networks with genetic algorithms; Part III Revisiting Modularity; 7 Evaluation of CALM; 7.1 The status of CALM; 7.2 Biological plausibility; 7.3 Computational plausibility; 7.4 Psychological plausibility; 7.5 Conclusions; Appendix A; Appendix B1 Hardware and software for neural networks; Appendix B2 Virtual implementations on transputer networks; Appendix B3 Hybrid implementation: the BSP400, a dedicated multiprocessor |
|
Appendix B4 Physical implementation: some notes on CALM in analog hardwareAppendix B5 Modular neurosimulators; Bibliography; Name index; Subject index |
| Summary |
This book introduces a new neural network model called CALM, for categorization and learning in neural networks. The author demonstrates how this model can learn the word superiority effect for letter recognition, and discusses a series of studies that simulate experiments in implicit and explicit memory, involving normal and amnesic patients. Pathological, but psychologically accurate, behavior is produced by ""lesioning"" the arousal system of these models. A concise introduction to genetic algorithms, a new computing method based on the biological metaphor of evolution, and a demonstration |
| Notes |
Print version record |
| Form |
Electronic book
|
| ISBN |
9781317781370 |
|
1317781376 |
|
