Naturally Intelligent Systems

Caudill, Maureen; Butler, Charles

ISBN 10: 0262031566 / ISBN 13: 9780262031561

Published by MIT Press, Cambridge, MA, 1989

About this Item

Cloth.; Octavo.

Seller Inventory # 84879cx1

Bibliographic Details

• Title: Naturally Intelligent Systems
• Publisher: MIT Press, Cambridge, MA
• Publication Date: 1989
• Language: English
• ISBN 10: 0262031566
• ISBN 13: 9780262031561
• Binding: Hardcover
• Edition: First Edition.
• Condition: Near Fine
• Dust Jacket Condition: Near Fine
• Weight: 2 pounds
• Seller catalogs:
  • Biology

About this title

Synopsis

A detailed study of neural networks offers an informative look at the operation and uses of these systems, discussing their role in the development of artificial intelligence, as well as their applications in speech, vision, robotics, and pattern recognition

Reviews

Neural networks are information-processing computers whose ever-shifting pattern of activity resembles the workings of the human brain. Once hailed as the next great breakthrough in computer science, this nascent technology is now, the authors concede, a fragile sapling. Yet Caudill, a regular contributor to AI Expert , and Butler, a specialist in neural network development, point out that these brainlike devices have found practical uses in such areas as speech recognition, robotics and signal processing. Future applications might include vision systems, quality control and independent robots capable of autonomous learning. This accessible if demanding primer leads the reader gradually from basic principles to techniques and applications. A sprinkling of cartoons, diagrams and anecdotes lightens the presentation.
Copyright 1989 Reed Business Information, Inc.

The "naturally intelligent systems," also known as neural networks, neurocomputers, or neural computers, are the newest generaton of computer systems. Their basic architecture is designed to imitate the structure of the brain by attempting to replicate some of its functions. Caudill and Butler cover the subject xlcomprehensively. There are detailed explanations on different system designs; how these systems "learn"; and what types of tasks they can handle. However, despite the fact that the book has a glossary and does not carry a single mathematical formula, numerous ideas are expressed with mathematical, computer science, and electrical engineering terminologies, which can be hurdles to readers without the appropriate background. Nonetheless, this is one of a very few books (if not the only one) on neural networks that is written for the layperson. This is a suitable addition to undergraduate collections, but is not recommended for public libraries.
- May Rathbone, Univ. of Washington Libs., Seattle
Copyright 1990 Reed Business Information, Inc.