Looking for good book on Numerical Methods and/or Optimization

[hotvette]

Any recommendations? The books I have are very outdated. Extremely important to me are:

- worked examples #1 criteria. Need that bridge between theory and implementation.

- not overly heavy on theory (don't want to hire a PhD to explain it). I have an MS Engineering level education (PDE's was the last math class I took).

- would love the book to include Singular Value Decomposition (SVD) and Optimization techniques, though I have a feeling that a good treatment of Optimization would need to be in a separate book

Appreciate any/all suggestions. Thanks!

 

[Dr Transport]

Schaums outline isn't too bad, fits criteria #1 very well.

 

[hotvette]

OMG, I had totally forgotten about Schaums. Seems like a lifetime ago (sorta was, actually). On the way courtesy of Amazon. Thanks.

In the meantime, I'd like to re-phrase my question and request additional replies.

Are there any Numerical Methods or Optimization books that you absolutely love (or like alot)? If, so I'd appreciate the title/author/edition and comments as to why.

Thanks

 

[PerennialII]

Keeping in mind #1 & and not going 'too deep' in math how about "Practical Optimization Methods" by Bhatti. It's pretty recent and the whole book is written with mathematica implementations in mind (contains a CD of the stuff)(similar books exist if going to use for example MATLAB as a platform, but don't have experience about those). What I like about it in particular is that it presents recent methods for a wide variety of different numerical optimization problem 'categories' and it's heavy with respect to the implementation aspect (and tolerable if you like to avoid "excess" math).

 

[neurocomp2003]

Numerical Recipes in C/ Numerical Recipes in C++/Numerical Recipes in Fortran/ NUmerical Methods in Matlab

 

[hotvette]

Dr Transport, PerennialII, neurocomp2003,

Thanks for your suggestions.

Keep the recommendations coming...

Any 'classics' come to mind (like Schlichting is to Boundary Layer Theory, Zienkiewicz is to the Finite Element Method, Timoshenko is to Theory of Plates and Shells, Knuth is to Computer Algorithms, Kernighan and Ritchie are to the C Programming Language, etc.)?

 

[LeBrad]

I know I'm almost two months late, but I can't resist a question about numerical method books!

For optimization, the best book I have seen is free right here
http://www.stanford.edu/~boyd/cvxbook/
Lots of examples in there.

For numerical methods, I have never seen a good book on solving ODEs and stuff like FEM, but a wonderful book on the important smaller numerical techniques (like system solving and SVD) is
http://web.comlab.ox.ac.uk/oucl/work/nick.trefethen/text.html
That book is very readable, and presents things in a very intuitive manner. Definitely the best book I have seen on the subject.

One other thing I have encountered that I thought gave very intuitive explanations is
http://www.cs.cmu.edu/~quake-papers/painless-conjugate-gradient.pdf
It basically explains how iterative system solvers work, and has lots of pictures to help explain things.