Neural Network Design (Electrical Engineering) [Hardcover]

Beale Hagan Demuth (Author), Mark Beale (Author), Martin T. Hagan (Author)

Book Description

Series: Electrical Engineering | Publication Date: December 29, 1995

The well-known, respected authors who developed the Neural Networks toolbox and the Fuzzy Systems Toolbox now bring you this text designed for electrical and computer engineering or computer science beginners. The book covers neuron model and network architectures, signal and weight vector spaces, linear transformations for neural networks. and performance surfaces and optimum points.

Editorial Reviews

About the Author

Ph.D. Stanford Univeristy

Product Details

• Hardcover: 736 pages
• Publisher: PWS Pub. Co.; Har/Dis edition (December 29, 1995)
• Language: English
• ISBN-10: 0534943322
• ISBN-13: 978-0534943325
• Product Dimensions: 9.3 x 8.2 x 1.4 inches
• Shipping Weight: 3.1 pounds
• Average Customer Review: 4.8 out of 5 stars  See all reviews (9 customer reviews)
• Amazon Best Sellers Rank: #425,379 in Books (See Top 100 in Books)
  • #62 in Books > Computers & Technology > Computer Science > Artificial Intelligence > Neural Networks

More About the Author

Discover books, learn about writers, read author blogs, and more.

Customer Reviews

Most Helpful Customer Reviews

41 of 41 people found the following review helpful:

4.0 out of 5 stars Excellent intro to NN maths but few practical advices, November 5, 1999

By 

Hans Ivers ([e-mail]) (Quebec, Canada) - See all my reviews

This review is from: Neural Network Design (Electrical Engineering) (Hardcover)

I read the entire book over a one-semester graduate course in NN. I was amazed by the quality of formalism (notation), which allow me to understand quite easily complex mathematical concepts, algorithms and proofs presented throughout the book. Authors introduced in an effective way all important mathematical concepts before using them. I felt this book is accessible for a beginner in NN field but you will need a good basis (one or more undergraduate courses) in linear algebra and calculus. Overall, this book constitutes an excellent introduction to NN but you will need an additional book to help you through more practical aspects of NN training. My suggestions are Chris Bishop (1995) Neural Networks for Pattern Recognition (chap. 8-9). or Reed & al. (1999). Neural Smithing : Supervised Learning in Feedforward Artificial Neural Networks.

16 of 16 people found the following review helpful:

5.0 out of 5 stars Excellent book for understanding neural network innards., June 29, 1999

By A Customer

This review is from: Neural Network Design (Electrical Engineering) (Hardcover)

I took a graduate neural networks course with Dr. Hagan who used this book. The book analyzes the contemporary algorithms for neural nets and shows why neural nets work (and don't work). MATLAB examples are on the supplemental disk but they can be coded easily in other languages. The convergence toward a solution is shown using 2D and 3D plots.

13 of 13 people found the following review helpful:

5.0 out of 5 stars Hands down the best introduction, January 19, 2004

By 

Louis Charbonneau (Montreal) - See all my reviews
(REAL NAME)   

This review is from: Neural Network Design (Electrical Engineering) (Hardcover)

I knew the very poor Matlab Neural Network Toolbox User's Guide by the same authors and I was kind of expecting the same, and boy was I wrong!

This book is simply brilliant, a miracle of pedagogy. It is intended for undergrad classes, but it is so clear that graduate students will benefit enormously from reading it before any other material. Plainly put, this book makes you UNDERSTAND this difficult topic, more than any other book that I know of (Zurada, Smith, Hassoun, Haykin, Duda-Hart, Caudill, etc)

A selection of worked out problems are included at the end of each chapter, a practice that is highly beneficial but alas too rare in books of the kind.

I very much appreciated the very clear exposition of backpropagation, and optimization methods such as Levenberg-Marquardt.

A note to Matlab users: funky demos are available for free and illustrate the main points of the book.