Simulation Books

I'm a scientist in natural science and I always wanted to gain more out of my data by use of some mathematics. However before I was very reluctant and shy to even touch the subject of modelling, mainly due to the fact that I was afraid of advanced mathematics. However, the material presented in the book is so easy to understand and can readily be absorbed. Especially the combination between the mathematics and the solutions via a modern modelling package I found very attractive.

I will continue using the material presented in the book for my own research and can recommend it to anyone who is interested to gain a basic understanding of models in the soil-biosphere.

english

I found this book to be really informative. I was looking for something to help decipher the scientific literature around molecular dynamics and protein folding and this book fit the bill perfectly.

The only other comment I have is that the 'for beginners' part of the title may be a little misleading. Someone approaching this book should have a backround in organic chemistry, physics, and some solid math (linear algebra and multivariate calculus) to really enjoy it.

This book provides a good introduction into Monte Carlo simulation for so-called microbeam analysis.

I read this book to start up my work on image formation of scanning electron microscopy. This book was really useful to understand how the Monte Carlo technique could be realized in my lap-top.
Although the author provides a complete set of PASCAL codes,
it was not difficult to convert them into C/C++ based my programming codes.
This book contains many references, mainly journal papers.
So advanced modificaion may require ones to read these articles.
But without this kind of text book,
the analysis by Monte Carlo technique would be more difficult.

To understand the physical background of electron transport in a specimen,
I recommend to read SEM text books by Reimer or by Goldstein et al.

In the field of Silicon Processing, there hasn't been a text book for engineers who want to get information of advanced technologies.

The book is a great reference for some of the advanced features of COMSOL. Also it includes state of art case steadies in different fields. It allows COMSOLs' users to go deeper in the analysis.

I took the class at the University of MI when he was writing the book. It was very interesting, and the Maple examples helped me tremendously in visualizing the new concpts introduced in the course. Multivariable Calculus is a difficult subject, and this book goes a long way towards simplifying it and making it accessible.

This book is an excellent introduction to neural networks for the reader with a good mathematics background. The first few chapters describe popular and useful algorithms for implementing neural networks (Hopfield, backpropagation). Much of the book is devoted to the theory behind these networks (capacity, convergence and the dynamical systems approach, Markov chains). [This gave me, a total novice before reading the book, a much better idea of how these networks function, so much so that I can implement them in software.] Complementing the theory are numerous stimulating projects that illustrate many applications and open questions in the field.

De Wilde's book is very clearly written. Its compact size is a joy in this age of mammoth, but often poorly-written and substanceless, textbooks.

This book provides a detailed, yet accessible technical explanation of the inner workings of backpropagation of errors, the most commonly used training method for feedforward neural networks.

This book is for the programmer who wishes to implement backprop, or for the data miner who wishes to understand backprop. This book is not well-suited for people without a technical (math) background.

Pros:
-Pseudocode is provided
-Explanation advances from plain vanilla backprop to variants including enhancements like adaptive step size and momentum
-Application of neural networks is covered, including practical issues such as appropriate data representation and avoidance of over-fitting
-A few alternative architectures are explained (briefly) at the end of the book

Cons:
-This book was published in 1996, so it is getting a little old, though it has held up well

A variety of interesting articles concerning the neurobiology of neural networks. For example, one article by Gardner is "Static Determinants of Synaptic Strength". Although backpropagation as used in artificial neural networks is unlikely to be used by biological nervous systems, it is shown both in the buccal ganglia of Aplysia that synaptic strengths can be specified by postsynaptic neurons. As well, in the mammalian hippocampus, nitric oxide (NO) and arachidonic acid may serve as retrograde messengers from the postsynaptic cell for induction of long-term potentiation.