Customer Reviews

Evolutionary Computation

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

 

 

I have read some excellent books and survey papers describing particular domains of the evolutionary computation field, but this book try to give us a "Unified Approach" of this field.
I find very interesting the author's idea to model the evolutionary process with a dynamic (nonlinear!)system so that "...given particular initial conditions the system follows a trajectory over time through a complex evolutionary state space. One can than study various aspects of these processes such as their convergence properties, their sensitivity to initial conditions, their transient behavior and so on."[pg. 2 of the book]. And, I can add, why not "chaotically behaviors".

Prof. Alexandru Serbanescu, Ph.D.,
IEEE Senior Member of CAS and Computational Intelligence Societies

De Jong shows how to apply the metaphor of biological evolution to many problems outside biology. He shows that in some other context, there are a few crucial issues to be determined. Firstly, what is an individual? Perhaps a particular algorithm, with certain parameter values. Then, what does it mean to have competing individuals? What is the system within which you are determining the "fitness" of an individual. Finally, how to successful individuals pass on their traits, and can these be mixed between such individuals? The latter is of course the analog of sexual reproduction.

The text describes evolutionary algorithms can be deployed as problem solvers, if you can settle the issues in the previous paragraph. These set the stage for the bulk of your computations.

This book shows a unified view of genetic algorithms, genetic programming, and related fields. It includes some discussion of the historical development of the disciplines and presents the theory behind them.

However, while it does fit an undergrad course, it is not a book for practitioners, since most algorithms are not explained in detail.