This is a great introduction to algorithms and computations in Statistical Mechanics. The book does not assume a lot of prior knowledge of Physics/Statistical Mechanics which is a great plus. This book is rare because it takes a very strong computational approach towards Physics (unlike most Physics textbooks that approach the subject with a very analytical approach with "nice" functions). The computational approach is really the approach you need to take when doing "real" Physics, in my opinion.
I also find that the book has a rare and sparse elegance too it: Its not needlessly verbose and the author explains his subject without drowning you in words. The approach is not to drown you in equations (either) but teach you just enough so that you understand the Computer Algorithm (which is the primary focus). There are a lot of interesting graphs and diagrams across the book too.
The algorithms explained in this book like the Metropolis algorithm, Simulated Annealing, Random Walks etc. have widespread application in areas outside Physics too. So I would recommend this book to people who find these algorithms abstract to take a look to get a good feel on how these algorithms work in a concrete setting. In other words, even if you are not a physicist, understanding how these algorithms work in physical systems will help in other areas like Machine Learning!
I am almost at the end of a MOOC on Statistical Mechanics taught by the author of this book. The MOOC is amazing (like the book). Ideally it would be amazing to read this book in conjunction with the course but this may not be possible for everyone, of course.
If I would offer a suggestion it would be the author should choose a particular computer language for the algorithms in this book in a future edition. Currently they are in psuedocode. Of course, this way these algorithms are more generic and can be implemented by interested readers in any language. But by implementing them in a particular language they become more "alive". In the MOOC course I'm taking, these algorithms are implemented in Python and there is a sort of instantaneous and interesting quality to them. You can run them, you can play around with them. This is not the case with the psuedocode in the book, sadly.
- Paperback: 360 pages
- Publisher: Oxford University Press UK; Pap/Cdr edition (1 August 2006)
- Language: English
- ISBN-10: 0198515367
- ISBN-13: 978-0198515364
- Product Dimensions: 24.4 x 19.1 x 1.9 cm
- Boxed-product Weight: 762 g
- Average Customer Review: Be the first to review this item
- Amazon Bestsellers Rank: 100,937 in Books (See Top 100 in Books)