Paradigms of AI Programming is the first text to teach advanced Common Lisp techniques in the context of building major AI systems. By reconstructing authentic, complex AI programs using state-of-the-art Common Lisp, the book teaches students and professionals how to build and debug robust practical programs, while demonstrating superior programming style and important AI concepts. The author strongly emphasizes the practical performance issues involved in writing real working programs of significant size. Chapters on troubleshooting and efficiency are included, along with a discussion of the fundamentals of object-oriented programming and a description of the main CLOS functions. This volume is an excellent text for a course on AI programming, a useful supplement for general AI courses and an indispensable reference for the professional programmer.
Peter Norvig gives us an amazing collection of Lisp programs solving real-world AI problems. I must confess I skimmed some of the more esoteric parts of the book, such as the natural language processing, or implemeting Prolog in Lisp, but other parts (especially the one about an Othello-playing program) I found deeply fascinating--especially as I've myself tried to write a game-playing program in Lisp (https://github.com/dlindelof/saikoro.git)
Although this book includes a primer on Lisp, it is by far not a book about learning Lisp. This is more about showing how modern AI problems can be solved with a truly generic programming language such as Lisp.
The book is ginormous, and reading it is not a light undertaking but is well worth the effort. Will you become a better programmer after reading it? Not necessarily, unless you happen to be involved in writing this kind of programs. Read The Little Schemer instead if you want to become a better person. But I dare say that reading this book might open your eyes to the possibilities offered by this amazing language.
There are parts of this book I would give 5 stars too, parts that only get a 2. I will give it 3.5 stars overall (4 I guess since I can't give .5). I read this right after the SICP book, which is a solid 5 stars. SICP was full of insight, history and head nodding confirmations of stuff I have believed for a long time about the structure of programs. The Knuth books also give me that warm fuzzy kind of programming euphoria that SICP does, I just didn't get as much of that from Paradigms of Artificial Intelligence. Lots of good stuff, a bit of which I actually used in a recent Clojure project, but I was just hoping for more of that special feeling I get from the other classics. Perhaps if I was more into AI it would be different. This book was more on par with Paul Graham's On Lisp, if you have read that.
I can't say enough good things about this book. Although Lisp is in the title, don't let that put you off. This book is really about logically structuring your code and clarity of reasoning. Working through the early AI examples such as Eliza is a thoroughly enjoyable journey of learning.
Intro into Common Lisp with overview of specific outdated use cases. This book does not cover high level difficulties in AI development, falling back to description of historically significant AI projects, that are not representative of the field anymore and are covered on a very basic level.
Most of the AI is GOFAI, i.e. mainly concerned with search and symbol manipulation. In my mind the best part of the book is the Lisp itself -- Norvig is a great programmer and writer.