Genetic Programming IV: Routine Human-Competitive Machine Intelligence

Rating:
Summary: Gp here we Go
Review: Since using evolutionary algorithms for my work( I am currently a Phd Student). It is easy to see how many of the current EA could be used to solve or tackle various real world problems. But what Koza does once again is to argue the case that GP is more than just an optimization algorithm but instead an algorithm that tries to show what AI should do and how user and AI should interact to solve a problem. Once again numerous examples are given, with detail to how the problem is laid out so as to get the best results from the GP. This is highly important especially when applying GP to areas where they have not being applied. Koza shows that with well thought out planning GP's can be applied to all sorts of fileds. In one of the chapters he also lays some of the characteristics a problems should have for GP to be applicable.
Alround Great work, my advice get all his books and digest how he approaches various problems with GP. This example format Koza uses is far more useful than talking about what GP is and its theory. Though for a good intro into Evolutionary Algorithms including GP get either Foundations of Genetic Programming or an Introduction to Genetic Programming. An all round good intro is Introduction to Evolutionary Computing.