zbMATH — the first resource for mathematics

Neural networks. An introduction. 2nd upd. a. corr. ed. (English) Zbl 0831.68083
Berlin: Springer-Verlag. xvi, 330 p. (1995).
Since the publication of the first edition (1990; Zbl 0727.68002), the number of studies of the properties and applications of neutral networks has grown considerably. The field is now so diverse that a complete survey of its recent developments is impossible within the scope of an introductory text. Fortunately, a wealth of material has become available in the public domain via the Internet and the World Wide Web (WWW). We trust that the reader who is interested in some of the latest developments will explore this source of information on their own. We have, therefore, resisted attempting to completely update the text and references with the latest developments and rather concentrated our efforts on eliminating two shortcomings of the first edition. We have added a separate chapter on applying genetic algorithms to neural-network learning, complete with a new demonstration program neurogen. A new section on back-propagation for recurrent networks has been added along with a new demonstration program btt which illustrates back-propagation through time. In addition, we have expanded the chapter on applications of neural networks, covering recent developments in pattern recognition and time series prediction.

68T05 Learning and adaptive systems in artificial intelligence
68T10 Pattern recognition, speech recognition
92B20 Neural networks for/in biological studies, artificial life and related topics
68-01 Introductory exposition (textbooks, tutorial papers, etc.) pertaining to computer science
Full Text: DOI