Customer Reviews

Simulating Neural Networks with Mathematica

5 star:		(1)
4 star:		(1)
3 star:		(0)
2 star:		(0)
1 star:		(0)

 

 

This is another book where the capabilities of Mathematica are put to good use. Clear explanations and code make it a joy to go through and do all the calculational stuff. Helps even quite experienced people to visualise some of the concepts they may not be experienced with. All the basic models are dealt with. The last chapter on genetic algorithms is a bonus.

Overall, I am quite happy with the book. It does exactly as it describes...shows the reader how to use mathematica to simulate several types of Neural Networks. The code is clear, fairly short and the example networks fun to work though. The flexibilty of Mathematica made it a simple task to view what the networks were doing and thus made the networks easier to understand.

My only complaint is that the book is too short. Part of this complaint is that I really enjoyed playing with the example nets and hated to see it end. However, only about 8 networks are mentioned and each is covered in 20-30 pages...program code included. I wish the author had time to double or triple the size of the book to cover the nets more thoroughly and to cover others.

The book is also a bit shy on the mathematical treatment of the networks. It does have some math, but the derivations included are not that rigorous. I have supplemented this book with others to cover the theory.

Nevertheless, If you're wanting to use Mathematica for Neural Networking and you're having difficulty getting started, this book is worth the price.