Neural Smithing: Supervised
Learning in Feedforward Artificial
Neural Networks by Robert J. Marks II
Saves You Months Of Information Gathering
A practical book, Neural Smithing is aimed at the reader who intends to
design and build neural networks for applications from forecasting to
pattern recognition. The authors concentrate on multilayer perceptrons
(MLPs) as the most commonly used neural network model, which adds to
the books overall clarity and focus. This textbook-style reference begins
with simple, single-layer networks and the elements of supervised
learning. It then builds on these basics with such topics as error surfaces,
genetic algorithms, and generalization. Examples and illustrations guide
the reader through the discussion, but the authors dont suggest problems
for further study--a small omission in an otherwise well-constructed book.
Readers must know calculus and statistics to make sense of the text, but
they dont need much knowledge of neural computing. Whether used as
an introductory textbook or as a professional reference, Neural Smithing is

highly useful. Tightly focused and easy to use, it should have a place next
to every neural toolbox. --Rob Lightner
Personal Review: Neural Smithing: Supervised Learning in
Feedforward Artificial Neural Networks by Robert J. Marks II
I bought this 9 or 10 years ago when I became interested in AI, but my
math skills were pretty weak at the time. I've since completed a CS &
math degree, and I find this book easy to read, thoroughly interesting, and
insightful.
Many of the questions I found myself asking while I read were soon
answered as I read later sections of the book.
For those considering the purchase and unsure whether they can handle
it, for much of the book a decent exposure to calculus will suffice. For a
few chapters some exposure to ordinary differential equations would be
wise. Numerical Analysis is probably a good idea as well.
An exposure to probability & statistics (not the freshman version) would
help as well. The section on initialization techniques talks about various
probability distributions when determining methods for initializing weights.
If you don't care about the why's, it can be used as a reference for coming
up with a scheme for weight initialization, but I find it handy to know why
my code is doing something so I better know how to tweak it.
-Brian
For More 5 Star Customer Reviews and Lowest Price:
Neural Smithing: Supervised Learning in Feedforward Artificial Neural Networks by
Robert J. Marks II 5 Star Customer Reviews and Lowest Price!