Concepts for Neural Networks: A Survey

ISBN-13: 9783540761631 / Angielski / Miękka / 1997 / 312 str.

L. G. Landau; Lawrence J. Landau; John G. Taylor

Opis

A survey of concepts for neural networks. It discusses issues such as: the mathematics of neural networks; neurobiological modelling; neural network control of a simple mobile robot; attentional modulation in visual pathways; and the modelling and understanding of consciousness.

Spis treści

1 Neural Networks: An Overview.- 1.1 Introduction.- 1.2 What is a Neural Network?.- 1.3 Neural Network Dynamics.- 1.4 Training a Neural Network.- 1.5 Problems for Artificial Neural Networks.- 1.6 Problems for Neurobiologically Realistic Neural Networks.- 1.7 Conclusions.- 1.8 Further Reading.- 1.9 Bibliography.- 2 A Beginner’s Guide to the Mathematics of Neural Networks.- 2.1 Introduction: Neural Information Processing.- 2.2 From Biology to Mathematical Models.- 2.2.1 From Biological Neurons to Model Neurons.- 2.2.2 Universality of Model Neurons.- 2.2.3 Directions and Strategies.- 2.3 Neural Networks as Associative Memories.- 2.3.1 Recipes for Storing Patterns and Pattern Sequences.- 2.3.2 Symmetric Networks: The Energy Picture.- 2.3.3 Solving Models of Noisy Attractor Networks.- 2.4 Creating Maps of the Outside World.- 2.4.1 Map Formation through Competitive Learning.- 2.4.2 Solving Models of Map Formation.- 2.5 Learning a Rule from an Expert.- 2.5.1 Perceptrons.- 2.5.2 Multi-Layer Networks.- 2.5.3 Calculating what is Achievable.- 2.5.4 Solving the Dynamics of Learning for Perceptrons.- 2.6 Puzzling Mathematics.- 2.6.1 Complexity due to Frustration, Disorder and Plasticity.- 2.6.2 The World of Replica Theory.- 2.7 Further Reading.- 2.8 Bibliography.- 3 Neurobiological Modelling.- 3.1 Introduction.- 3.2 The Single Nerve Cell.- 3.3 Retinal Processing.- 3.4 The Cortical Streams.- 3.5 The Ventral Stream.- 3.6 The Dorsal Stream.- 3.7 Object Construction.- 3.8 Modelling the Processing Streams.- 3.9 Conclusions.- 3.10 Bibliography.- 4 Neural Network Control of a Simple Mobile Robot.- 4.1 Introduction.- 4.2 The Robot ‘Insects’.- 4.3 Neural Network for Obstacle Avoidance.- 4.3.1 Learning Strategy.- 4.3.2 Results.- 4.4 A Compound Eye.- 4.4.1 The First Compound Eye.- 4.4.2 Estimating Robot Position.- 4.4.3 Weightless Networks.- 4.4.4 Multi Discriminator Network.- 4.4.5 Processing Grey Level Data.- 4.4.6 Weightless Networks and the First Eye.- 4.4.7 Results.- 4.4.8 The Second Eye.- 4.5 Navigation.- 4.6 Discussion.- 4.7 Bibliography.- 5 A Connectionist Approach to Spatial Memory and Planning.- 5.1 Overview.- 5.2 Introduction.- 5.3 Biological Spatial Memory.- 5.3.1 Stimulus-Response Theory.- 5.3.2 Cognitive Maps.- 5.3.3 Topological Network-Maps.- 5.3.4 Planning.- 5.3.5 Summary.- 5.4 Connectionist Implementation.- 5.4.1 Principles of a Network-Map Based Artificial Spatial Memory.- 5.4.2 Sparse Versus Distributed Representation of Views.- 5.4.3 Implementation of a Sparse Model.- 5.4.4 A Constructive Learning Procedure for States and Actions.- 5.4.5 Planning and Search Procedure.- 5.5 An Experiment with a Robot.- 5.5.1 Task and Experimental Set-up.- 5.5.2 Learning and Planning Test.- 5.6 Discussion.- 5.7 Conclusion.- 5.8 Appendix: Vision system for letter detection and recognition..- 5.9 Appendix: Vision system for state recognition.- 5.10 Appendix: Notes on a biologically plausible implementation.- 5.11 Bibliography.- 6 Turing’s Philosophical Error?.- 6.1 Turing’s Machine.- 6.2 Digital Problems.- 6.3 Turing’s Thesis.- 6.4 How Turing Built his Universal Machine.- 6.5 Gödel’s Attempt to Break Turing’s Thesis.- 6.6 Yes-or-No Questions.- 6.7 Material and Formal Proofs.- 6.8 The Dialectica Lecture.- 6.9 A Calculus of Concepts?.- 6.10 Turing’s Error.- 6.11 Further Reading.- 6.12 Bibliography.- 7 Penrose’s Philosophical Error.- 7.1 Can Computers Think?.- 7.2 Solving Problems in Arithmetic.- 7.3 The Barber Paradox.- 7.4 Gödel’s Theorem.- 7.5 Penrose.- 7.6 A Computational Model for Thought?.- 7.7 Can Computers Think?.- 7.8 Appendix: Filling in Some Details.- 7.9 Bibliography.- 8 Attentional Modulation in Visual Pathways.- 8.1 Introduction.- 8.2 Structural Equation Modelling.- 8.3 Design and Image Acquisition.- 8.4 Image Analysis and Categorical Comparisons.- 8.5 Modelling of the Posterior Visual Pathway.- 8.6 Modelling Modulation by Interaction Terms.- 8.7 Interaction Effects Demonstrated with Regression Analysis.- 8.8 Regional Specificity.- 8.9 Conclusion.- 8.10 Bibliography.- 9 Neural Networks and the Mind.- 9.1 Introduction.- 9.2 The Relational Mind.- 9.3 Exploring the Relational Mind Model.- 9.4 Memory in the Relational Mind Model.- 9.5 A Neural Candidate for Global Control.- 9.6 Consciousness and Sites of Working Memory.- 9.6.1 Psychological Bases for Working Memory.- 9.6.2 PET Studies of Working Memory.- 9.7 Awareness of Words.- 9.8 Activity ‘Bubbles’ in Cortex.- 9.8.1 General Discussion.- 9.8.2 Technical Results.- 9.9 The Emergence of Qualia.- 9.10 Discussion.- 9.11 Bibliography.- 10 Confusions about Consciousness.- 10.1 Consciousness Studies.- 10.2 Bibliography.- 11 Round Table Discussion.- 11.1 Presentations.- 11.2 Audience Participation.