Название: Wavelets in Soft Computing, Second Edition
Автор: Marc Thuillard
Издательство: World Scientific Publishing
Серия: World Scientific Series in Robotics and Intelligent Systems
Год: 2023
Страниц: 320
Язык: английский
Формат: pdf (true), epub
Размер: 38.0 MB

The comprehensive compendium furnishes a quick and efficient entry point to many multiresolution techniques and facilitates the transition from an idea into a real project. It focuses on methods combining several soft computing techniques (fuzzy logic, neural networks, genetic algorithms) in a multiresolution framework. Illustrated with numerous vivid examples, this useful volume gives the reader the necessary theoretical background to decide which methods suit his/her needs. New materials and applications for multiresolution analysis are added, including notable research topics such as Deep Learning, graphs, and network analysis.

The main goal of Wavelets in Soft Computing is to furnish a synthesis on the state of integration of wavelet theory into soft computing. Wavelet methods in soft computing can be classified into five main categories that form the backbone of the book:

- Preprocessing methods
- Automatic generation of a fuzzy system from data
- Wavelet networks
- Wavelet-based nonparametric estimation and regression techniques
- Multiresolution genetic algorithms and search methods.

The main new contributions of Wavelets in Soft Computing to these topics are the automatic generation of a fuzzy system from data (fuzzy wavelet, fuzzy wavenets for online learning), wavelet networks, and wavelet estimators (extension to biorthogonal wavelets) and multiresolution search methods. These new methods have been implemented in commercial fire detectors or used during development. Although over 2000 articles have combined elements of wavelet theory to soft computing, no book has been dedicated to that topic yet. The topic has grown to such proportions that it is not possible anymore to offer an exhaustive review. For that reason, the emphasis is placed on topics that are not specific to a particular application. A special place is given to methods that have been implemented in real-world applications. It is especially the case of the different techniques combining fuzzy logic and neural networks to wavelet theory. These methods have been implemented during the development of several products and have found applications in intelligent systems, such as fire detection.

Another example is the field of Deep Learning. Many Deep Learning applications incorporate a wavelet decomposition stage to better capture features at different resolutions, a quite sensible step as the size of an object in an image may greatly vary. A fascinating aspect that we discuss in a new chapter is that multiresolution is at the heart of the functioning of Deep Learning.

Neural networks on graphs are important in studying communication networks and analyzing internet data. Here also, multiresolution permits a better analysis. The research community has broadly integrated the idea that the integration of multiresolution often improves algorithms. This new edition aims to capture some of these exciting new developments.

Readership: Researchers, professionals, academics and graduate students in fuzzy logic.