Knowledge-based neurocomputing: a fuzzy logic approach
Fuzzy logic; logic of vagueness (03B52) Research exposition (monographs, survey articles) pertaining to computer science (68-02) Learning and adaptive systems in artificial intelligence (68T05) Problem solving in the context of artificial intelligence (heuristics, search strategies, etc.) (68T20) Logic in artificial intelligence (68T27) Knowledge representation (68T30) Theory of languages and software systems (knowledge-based systems, expert systems, etc.) for artificial intelligence (68T35)
This book is part of the series ``Studies in Fuzziness and Soft Computing. It introduces a novel approach to building fuzzy rule-bases, called Fuzzy All-permutations Rule-Bases (FARB). The book starts with a broad introduction to the field of connectionism, namely artificial neural networks, fuzzy rule-bases, and neurocomputing, before it introduces the FARB concept. The authors show that inferring the FARB using standard tools from fuzzy logic theory yields an input-output relation that is mathematically equivalent to that of standard neural networks. Chapter 2 formally defines the FARB approach and shows that its special structure implies a closed-form formula for its input-output mapping. Some special cases of the FARB are discussed in Chapter 3, where mathematical equivalence between these cases and various types of artificial neural networks is discussed. Example applications are presented to illustrate the main ideas. Since FARB may include either a large number of rules or complicated rules which may obstruct its comprehensibility, Chapter 4 presents a systematic procedure for rule reduction and simplification to minimize the effect of this. Application of FARB to large-scale problems is discussed in Chapters 5 and 6, where FARB is used for knowledge extraction from artificial neural networks trained to solve several benchmark problems. New knowledge-based design methods are described and their usefulness is demonstrated by designing networks to solve language recognition problems. Future research is summarized in Chapter 7, where potential applications based on transferring concepts and ideas between fuzzy logic and networks are discussed.
- Extracting symbolic knowledge from recurrent neural networks -- a fuzzy logic approach
- Mathematical modeling of the lambda switch: a fuzzy logic approach
- Knowledge extraction from support vector machines: a fuzzy logic approach
- Hysteresis nonlinearity identification using new Preisach model-based artificial neural network approach
- Fuzzy rule-based inference. Advances and applications in reasoning with approximate knowledge interpolation
- Fuzzy hyperbolic neural network with time-varying delays
- Analysis of artificial neural network learning near temporary minima: a fuzzy logic approach
This page was built for publication: Knowledge-based neurocomputing: a fuzzy logic approach
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q950498)
