An Introduction to the Modeling of Neutral Networks
bibliographyCHIR algorithmconstraints on synaptical weightsHebbian models of associative memoryinternal representation problemsLinkser ontogenetic modelneurocomputationself-organizationtemporal sequences of patternsternary synapses
Learning and adaptive systems in artificial intelligence (68T05) Pattern recognition, speech recognition (68T10) Introductory exposition (textbooks, tutorial papers, etc.) pertaining to biology (92-01) Neural networks for/in biological studies, artificial life and related topics (92B20) Neural biology (92C20)
Contents (brackets indicating approximate number of pages): 1. Introduction (10); 2. The biology of neural networks (45); 3. The dynamics of n.n.: a stochastic approach (40); 4. Hebbian models of associative memory (55); 5. Temporal sequences of patterns (20); 6. The problem of learning in n.n. (35); 7. Learning dynamics in ``visible n.n. (60); 8. Solving the problem of credit assignment (30); 9. Self- organization (25); 10. Neurocomputation (45); 11. Neurocomputers (30); 12. A critical view of the modeling of n.n. (20); References (40); Index (6). An extensive bibliography is given, the subject index is relatively small and, unfortunately, no notation list attached. Two levels of presentation are used; supplementary parts requiring more mathematics are distinguished by smaller fonts. The book appears in the collection: Monographs and Texts in Statistical Physics, which in itself gives certain characteristics of the approach adopted; the text freely uses rather complicated notation and some not explicitly explained conventions of the language of physics. Moreover, it sometimes adds new synonyms for established concepts, it is slightly repetitious and is not systematically ordered. In comparison with the similarly oriented book by \textit{J. Hertz}, \textit{A. Krogh} and \textit{R. Palmer} [Introduction to the theory of neural computation (1990)] the reviewed one is considerably less lucid. In some chapters a loose mixture of computer room, biochemistry lab and neurobiology department terms go hand by hand (e.g. in chapter 2 the reader will meet chemical formulas of dopamine, GABA and other neurotransmitters (what is this good for?), architecture scheme of peripheral structure including nigro-striate formation, cerebellum and hippocampus, experimental data on Aplysia, mean square deviation of potential defined by double integral --- all this without too much explanation). The author declares that the book is written so that it can be understood by a graduate in biology endowed with a little knowledge of algebra (?!); the author must have been really happy to meet such students. A diversity of topics is involved: a lot of philosophy, views, hypothetical constructions, explanations of the style of explanation. On the other hand, the same features which probably make the reading problematic, give the book a flavor of universality which only few authors can afford. The book brings thorough discussions of rich material, some of which can be otherwise hardly found even in present day textbooks [it should be taken into account that the text originated in 1988 as proceedings based on personally delivered lectures], like temporal dynamics using ternary synapses, internal representation problems reflected in the CHIR algorithm, constraints on synaptical weights, Linsker ontogenetic model etc. Certainly, not all often used paradigms are covered: there is no part dealing with the ART model, hidden neurons minimization techniques (say pruning), variants of response functions (say RBF), soundly based generalization theory etc. Summarizing, we offer a recommendation to the reader, who is not yet acquaintant with the main ideas conveyed by the book: Start with something else but do not forget then to go carefully through this introduction: you will surely find a lot of valuable and interesting ideas, examples, experimental data, formal techniques, original insights etc., which will pay off the time spent.
- Introduction to neural network models
- scientific article; zbMATH DE number 833421
- Models of neural networks
- Neural networks. An introduction
- An introduction to the mathematical theory of neural networks
- scientific article; zbMATH DE number 4121807
- scientific article; zbMATH DE number 825004
- scientific article; zbMATH DE number 867339
- Neural network models. An analysis
- A review of combinatorial problems arising in feedforward neural network design
- Feature space mapping as a universal adaptive system
- Anti-synchronization control for delayed memristor-based distributed parameter NNs with mixed boundary conditions
- Neural networks retrieving Boolean patterns in a sea of Gaussian ones
- Non-convex multi-species Hopfield models
- Neural network under external stimulus: improving storage capacity and reactions
- ANN approach to sorption hysteresis within a coupled hygro-thermo-mechanical FE analysis
- Maximum Memory Capacity on Neural Networks with Short-Term Synaptic Depression and Facilitation
- scientific article; zbMATH DE number 4049448 (Why is no real title available?)
- Connectionistic problem solving: computational aspects of biological learning
- Modeling Brain Function
- Models of neural networks
- scientific article; zbMATH DE number 51422 (Why is no real title available?)
- A quantization-model of the neural nets
- scientific article; zbMATH DE number 578246 (Why is no real title available?)
- scientific article; zbMATH DE number 4121807 (Why is no real title available?)
- scientific article; zbMATH DE number 825004 (Why is no real title available?)
- Routing control of packet flow using neural network
- The parallel updating in a fully-frustrated Ising model: a damage-spreading analysis
This page was built for publication: An Introduction to the Modeling of Neutral Networks
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4693709)